Methods and systems for creating a digital interconnect fabric

The invention relates generally to systems and methods for providing optical communication networks and, more specifically, to systems and methods for providing a digital interconnect fabric for optical communication networks.

Various network topologies exist for enabling terminal equipment at one node to communicate with terminal equipment at another node within the network. FIG. 1(A) illustrates a simple network comprised of a point-to-point connection between two pieces of terminal equipment T 1 and T 2 . In this network, terminal T 1 is able to send signals to T 2 and can receive signals from T 2 . Similarly, terminal T 2 can both transmit and receive signals from terminal T 1 . The link between terminals T 1 and T 2 may comprise a half duplex or full duplex line.

FIG. 1(B) is an example of an arbitrated loop connecting three or more terminals together in a network. In the arbitrated loop, terminal T 1 sends signals to terminal T 2 , terminal T 2 sends signals to terminal T 3 , terminal T 3 sends signals to terminal T 4 , and terminal T 4 sends signals to terminal T 1 . Thus, as shown in the diagram, terminal T 1 receives signals from terminal T 4 , terminal T 2 receives signals from terminal T 1 , terminal T 3 receives signals from terminal T 2 , and terminal T 4 receives signals from terminal T 3 . In essence, communication signals travel in one direction along the loop from a transmitting terminal T until it reaches a receiving terminal T. The terminals T in between do not process the signals but instead act as repeaters within the network. The arbitrated loop is an example of a ring network in which signals are passed from terminal to terminal until they reach the intended recipient or recipients.

FIG. 1(C) is an example of a network having a switch 10 . According to this type of network, the switch 10 enables connectivity between a set of M terminals and a set of N terminals. Each of the terminals T 1 to TM and T 1 to TN may transmit, receive, or both transmit and receive signals. The switch 10 is typically a cross-over switch for making the necessary connections between any one of the terminals T 1 to TM to any of the other terminals T 1 to TN.

FIG. 1(D) is an illustration of a typical hub network, such as one for Ethernet. With this type of network, a number of terminals are connected to each hub 12 . For instance, in this figure, terminal T( 1 )( 1 ) to terminal T( 1 )(M) are connected to a common hub 12 ( 1 ). Each of the terminals T( 1 )( 1 ) to terminal T( 1 )(M) communicate with each other through the hub 12 ( 1 ), which enables half duplex communication between the terminals. The hubs 12 may allow full duplex communication, in which case the hubs 12 may be considered switches. In either event, groups of terminals T communicate with each other through the hubs 12 . The hubs 12 are interconnected to each other through a backbone 14 to enable terminals T associated with one hub to communicate with terminals T at another hub.

All of the networks can be considered to have an interconnect fabric. The interconnect fabric generally refers to the ability of a network to direct communication signals from a terminal T at one node to a terminal T at another node within the network. For the network shown in FIG. 1(A) , the interconnect fabric may enable one of the terminals T to gain control of a common line which carries signals from either piece of terminal T to the other terminal T. For the network shown in FIG. 1(B) , the interconnect fabric may involve some type of token sharing whereby one of the terminals T is able to transmit signals along the loop or ring. For the network shown in FIG. 1(C) , the interconnect fabric refers to the switching of signals from one terminal T to another terminal T. For the network shown in FIG. 1(D) , the interconnect fabric refers not only to the interconnection between terminals T at one hub 12 but also the interconnection between terminals T at different hubs 12 .

Regardless of the network topology, the interconnect fabric also depends upon the communication protocol. One of the most common network protocols is the Ethernet, which is defined by IEEE Standard 802.3, which is incorporated herein by reference. Ethernet has evolved over the years and can be placed on different media. For example, thickwire can be used with 10Base5 networks, thin coax for 10Base2 networks, unshielded twisted pair for 10Base-T networks, and fibre optic for 10Base-FL, 100Base-FL, 1,000Base-FL, and 10,000Base-FL networks. The medium in part determines the maximum speed of the network, with a level 5 unshielded twisted pair supporting rates of up to 100 Mbps. Ethernet also supports different network topologies, including bus, star, point-to-point, and switched point-to-point configurations. The bus topology consists of nodes connected in series along a bus and can support 10Base5 or 10Base2 while a star or mixed star/bus topology can support 10Base-T, 10Base-FL, 100Base-FL, 1,000Base-FL, and Fast Ethernet.

Ethernet, as well as many other types of networks, is a shared medium and has rules for defining when nodes can send messages. With Ethernet, a node listens on the bus and, if it does not detect any message for a period of time, assumes that the bus is free and transmits its message. A major concern with Ethernet is ensuring that the message sent from any node is successfully received by the other nodes and does not collide with a message sent from another node. Each node must therefore listen on the bus for a collision between the message it sent and a message sent from another node and must be able to detect and recover from any such collision. A collision between message occurs rather frequently since two or more nodes may believe that the bus is free and begin transmitting. Collisions become more prevalent when the network has too many nodes contending for the bus and can dramatically slow the performance of the network.

Fibre Channel is another communications protocol that was designed to meet the ever increasing demand for high performance information transfer. As with Ethernet, Fibre Channel is able to run over various network topologies and can also be implemented on different media. For instance, Fibre Channel can work in a point-to-point network such as the one shown in FIG. 1(A) , in an arbitrated loop such as the one shown in FIG. 1(B) , and can also work in a cross-point switch configuration or hub network, such as those shown in FIGS. 1(C) and 1(D) . Fibre Channel is essentially a combination of data communication through a channel and data communication through a network. A channel provides a direct or switched point-to-point connection whereas a network supports interaction among an aggregation of distributed nodes and typically has a high overhead. Fibre Channel allows for an active intelligent interconnection scheme, called a Fabric, to connect devices. A Fibre Channel port provides a simple point-to-point connection between itself and the Fabric.

For most networks, including those that operate under Ethernet and Fibre Channel, the digital interconnect fabric requires some examination of the signals in order to provide the desired interconnection. For instance, in sending signals from one terminal to another, the arbitrated loop, switch, or hub must examine the address in order to ensure that the signal is delivered to the desired terminal. This overhead associated with the digital interconnect fabric is a burden on the network and generally decreases efficiency, speed, and overall performance of the network.

Some attempts have been made to improve performance by using optical communication. By using optical signals and fibers, electromagnetic interference (EMI), noise, and cross-talk can be substantially eliminated and transmission speeds can be increased. Even with optical communication, however, the digital interconnect fabric typically involves converting these optical signals into electrical signals. For instance, with the arbitrated loop, each terminal T receives optical signals, converts them into electrical signals, reviews the addressing information within the electrical signals, and either processes those signals if that terminal T is the intended recipient or regenerates optical signals and forwards them to the next terminal T in the loop. For the network shown in FIG. 1(C) , the switch 10 typically converts the optical signals into electrical signals in order to provide the desired interconnection between the terminals T. For the same reason, the hubs 12 also convert the optical signals from the terminals T into electrical signals in order to provide the proper routing to the desired destination terminal T. While the optical lines shield the signals from noise, cross-talk, and EMI, the need to convert the optical signals into electrical signals and then once again generate optical signals reduces signal quality, adds a layer of complexity and cost, and degrades the overall potential performance of the networks.

The invention addresses the problems above by providing systems and methods for providing a digital interconnect fabric between a plurality of nodes within a network. The network includes a bi-directional optical bus for routing digital optical signals between a plurality of nodes. Each node is connected to the bi-directional optical bus through a passive optical interface device. The passive optical interface device receives signals from the nodes and routes the signals in both directions along the bi-directional optical bus. The passive optical interface devices also take signals traveling along the bi-directional optical bus and route them to each node. Each of the nodes that transmits optical signals is assigned a unique wavelength of transmission. Because the network distributes optical signals from each node to every other node within the network, any node seeking to receive signals from another node can simply receive the signals at the wavelength corresponding to that node's transmission wavelength. The nodes that receive signals may detect signals from all wavelengths of transmission or may have a tunable receiver for selecting only a desired wavelength of transmission.

According to another aspect, in addition to having a wavelength of transmission for the optical signals, each transmitting node may also have the capability of transmitting control optical signals over a control wavelength. This control wavelength provides a control channel for establishing connections between any two nodes. The receiving nodes are able to detect the control signals at the control wavelength as well as optical signals at least one wavelength of transmission. The networks according to the invention do not require any hub or switch for converting optical signals into electrical signals in order to route the signals to the appropriate receiving node. Instead, the optical networks maintain the optical signals in the optical domain which results in improved signal quality, network efficiency, and faster transmission speeds.

Other advantages and features of the invention will be apparent from the description below, and from the accompanying papers forming this application.

The accompanying drawings, which are incorporated in and form a part of the specification, illustrate preferred embodiments of the present invention and, together with the description, disclose the principles of the invention. In the drawings:

FIGS. 1(A) to 1(D) are diagrams of different network topologies;

FIG. 2 is a diagram of an optical network according to a preferred embodiment of the invention;

FIG. 3 is a more detailed diagram of the optical network illustrating assignment of wavelengths to nodes.

Reference will now be made in detail to preferred embodiments of the invention, non-limiting examples of which are illustrated in the accompanying drawings. With reference to FIG. 2 , a network 20 includes a number of nodes 24 each having terminal equipment T. The nodes 24 are coupled to a bi-directional optical bus 21 through Optical Interface Devices (OIDs) 22 . The optical network 20 , as will be described in more detail below, provides a digital interconnect fabric that addresses many of the problems associated with conventional networks.

The OIDs 16 may comprise any suitable structure for directing optical signals from each node 24 onto the optical bus 21 in both directions and for directing optical signals traveling along the optical bus 21 in both directions toward each node 24 . Suitable OIDs are described in U.S. Pat. Nos. 5,898,801 and 5,901,260 and in co-pending patent application Ser. No. 10/280,967, entitled “Optical Interface Devices Having Balanced Amplification,” filed on Oct. 25, 2002, all of which are incorporated herein by reference.

The networks 20 according to the invention carry digital signals including, but not limited to, the Ethernet standard, as specified by International Standards Organization (ISO) 802.3, Mil-Std 1553 , ARINC-429, RS-232, NTSC, RS-170, RS-422, NTSC, PAL, SECAM, AMPS, PCS, TCP/IP, frame relay, ATM, Fibre Channel, SONET, WAP, VME, PCI, and InfiniBand.

Each node 24 has terminal equipment T that includes at least one of an optical-to-electrical converter and an electrical-to-optical converter. The electrical-to-optical and optical-to-electrical converters may be provided as part of an electro-optical interface circuit (EOIC) as described in U.S. Pat. Nos. 5,898,801 and 5,901,260. The invention is not limited to the type of optical transmitter but includes LEDs and lasers, both externally and directly modulated. As will be appreciated by those skilled in the art, each node 24 may also include translation logic devices and other devices used in the processing or routing of the signals as part of the terminal equipment T. A preferred network is described in U.S. Pat. No. 5,898,801 entitled “Optical Transport System,” which is incorporated herein by reference.

The optical bus 21 is preferably a single-mode fibre that carries optical signals in both directions simultaneously to all nodes 24 connected to the bus 21 . The optical bus 21 also preferably provides bi-directional optical amplification of the signals traveling along the bus, such as described in U.S. Pat. Nos. 5,898,801 and 5,901,260. Thus, the amplification of the optical signals may occur along a section 25 of the bus 21 interconnecting two of the nodes 24 . The optical amplification need not occur along these interconnection sections but alternatively may be provided along paths which interconnect the nodes 24 to the OIDs 22 . Furthermore, the optical amplification may occur within the nodes 24 or within the OIDs 22 . The optical amplification may be performed through fibre amplifiers, such as erbium-doped fibres or other rare-earth doped fibres, as described in U.S. Pat. Nos. 5,898,801 and 5,901,260. The amplification may also be performed by devices separate from the fibre, such as any of the various discrete laser amplifiers.

Significantly, the amplification that occurs within the network 20 associated with each node 24 compensates for splitting losses to and from that node 24 . In other words as optical signals travel down the bi-directional optical bus 21 and encounter an OID 22 , a fraction of the optical signals is diverted to the node 24 . To compensate for this loss in signal strength, the optical signals are amplified, such as up to their original level, to maintain signal quality and strength. Thus, when the signals arrive at the next downstream node 24 , the optical signals are at a level which can be received and processed by the node 24 . This process of diverting signals to each node 24 and amplifying the signals preferably continues at each node 24 . While each node 24 preferably has an associated amplifier, it should be understood that the amplifiers may not be associated with every node 24 but should be dispersed throughout the network so as to ensure sufficient signal strength for each node 24 . While optical amplifiers are preferably included within the network 20 to compensate for losses and to interconnect a greater number of nodes 24 , the networks according to the invention may employ no amplifiers or a fewer number of such amplifiers.

The nodes 24 can provide varying levels of communication functionality through their terminal equipment T. The nodes 24 may include only a receiver for detecting communications from the other nodes 24 and/or may have a transmitter for sending communications to the other nodes 24 . The nodes 24 may also include additional functionality, such as a display interface. Networks 20 according to the invention may include other numbers of nodes, may include additional or fewer types of nodes, and may include only one type of node. Additional details of the nodes 24 will become apparent from the description below.

The optical network 20 provides a number of advantages over existing systems that are installed in structures. For one, the nodes 24 communicate with each other through optical signals. Consequently, the network 20 enjoys immunity from electromagnetic noise whereby electrical systems within the terminal equipment T do not cause interference with normal operation of any one of the nodes 24 . Furthermore, the optical network 20 includes a single bi-directional bus 21 which can be used to interconnect a large number of nodes 24 . For example, the network 20 can accommodate in the range of 256 nodes 24 on the single fiber 21 . The network 20 therefore presents a viable solution for systems having more than eight to 10 components and, moreover, presents a single solution that can integrate multiple systems. Another advantage of the network 20 is that it greatly simplifies the amount of cabling associated with interconnecting nodes 24 . As mentioned above, the network 20 employs a single bi-directional bus 21 with every node 24 being connected to this one bus 21 through an OID 22 . This single bi-directional bus 21 greatly simplifies not only the installation of the network 20 but also the maintenance and repair of the network 20 .

A more detailed diagram of a network 30 according to the preferred embodiment of the invention will now be described with reference to FIG. 3 . The optical network 20 , as well as the network 30 , includes a digital interconnect fabric. With reference to FIG. 3 , the network 30 includes the bi-directional bus 21 , the OIDs 22 , and terminal equipment T. To highlight the advantages of the invention, FIG. 3 also illustrates groupings 32 of the terminal equipment and OIDs 22 which replace the conventional hub 12 shown in FIG. 1(D) . For instance, grouping 32 ( 1 ) is associated with terminal equipment T( 1 )( 1 ) to terminal T( 1 )(M) and the network 30 may include N number of additional groupings 32 . Thus, a grouping 32 (N) has terminal equipment T(N)( 1 ) to terminal T(N)(P). In general, the network 30 may include N number of groupings 32 with each grouping having one or more terminals T. In this example, grouping 32 ( 1 ) has M terminals while grouping 32 (N) has P terminals.

The digital interconnect fabric according to the preferred embodiment of the invention involves assigning each terminal T a unique wavelength for transmission. In the example shown in FIG. 3 , each terminal T has a different wavelength of transmission λ As mentioned above, some terminals T may be receive only in which case no wavelength of transmission needs to be assigned to that terminal T. Thus with reference to FIG. 3 , terminals T( 1 )( 1 ) to terminal T( 1 )(M) are assigned wavelengths λ (1)(1) to λ (1)(M) . Similarly, terminals T(N)( 1 ) to T(N)(P) are assigned wavelengths λ (N)(1) to λ (N)(P) . Each terminal T transmits at a unique wavelength and these signals are sent to the OID 22 and directed onto the bi-directional optical bus 21 in both directions. The signals from each terminal T thus travel along the bi-directional bus 21 and are routed to every other terminal T through the OIDs 22 . Thus, optical signals originating at any of the terminals T are routed to every other terminal T.

To receive signals from another terminal T, a receiving terminal T detects the optical signals at the wavelength corresponding to the transmitting nodes wavelength. According to one aspect, every receiving terminal T receives optical signals from all terminals T and converts all signals into electrical signals. According to this aspect, the receiving terminals T detect the signals from all transmitting terminals T. According to another aspect, the receiving terminals T have a tuneable receiver for selecting a desired wavelength of transmission so that the receiving terminal T can detect the signals from just one transmitting terminal T. Numerous ways exist for having a receiving terminal T tune into the wavelength of a desired transmitting terminal T. For example, each transmitting terminal T may also transmit control signals over a control wavelength λ C . Each receiving node detects any control signals at wavelength λ C with these control signals λ C coordinating the tuning of the receiving terminals wavelength to the wavelength of a desired transmitting terminal T. The receiving terminals T may also have the capability of transmitting at the control wavelength λ C for establishing channels between a transmitting terminal T and a receiving terminal T.

As should be apparent from the above description, the network 30 provides a digital interconnect fabric which, in essence, provides point-to-point connections between any two nodes 24 or terminals T within the network 30 . This digital interconnect fabric does not require any hubs or switches that convert optical signals into electrical signals for the purpose of routing the signals to the appropriate node. Instead, signals from all nodes 24 are available at every other node. By maintaining the signals in the optical domain, the networks according to the invention provide improved performance and improved signal quality.

The foregoing description of the preferred embodiments of the invention has been presented only for the purpose of illustration and description and is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations are possible in light of the above teaching.

The embodiments were chosen and described in order to explain the principles of the invention and their practical application so as to enable others skilled in the art to utilize the invention and various embodiments and with various modifications as are suited to the particular use contemplated.