The present invention generally relates to a tiered optical network architecture for transporting wavelength division multiplexed (WDM) traffic, and particularly relates to a higher-tiered optical network composed of peer network nodes that have reduced complexity.
Increasing the flexibility with which an optical transport network can route wavelength division multiplexed (WDM) traffic has traditionally increased the efficiency of the network. Reconfigurable optical add/drop multiplexers (ROADMs) have greatly contributed to this increased routing flexibility by enabling traffic at the wavelength granularity to be selectively added or dropped at any node in the network. However, ROADMs employ fairly complex and expensive components to provide this flexible routing capability, meaning that ROAMDs prove cost-prohibitive in some contexts.
One such context relates to a network that efficiently transports the traffic of multiple services in a converged fashion. Rather than employing multiple different networks in parallel for transporting these different services (e.g., mobile, business, and residential services), a converged network transports those services together using the same network. A transport network that optically converges different services by transporting those services on different wavelengths would be advantageous, for a variety of reasons, but has heretofore been precluded by the high cost of the necessary hardware components (e.g., ROADMs).
Consequently, known transport networks converge different services using packet aggregation instead. While packet aggregation currently requires less hardware expense for converged transport, that expense will not scale equally with the significant traffic increases expected in the near future. Moreover, while packet aggregation suffices in many respects for realizing convergence, it proves inefficient in implementation. Indeed, converging multiple services at the packet level involves significant complexity in order to accommodate the different packet requirements associated with the different services.
Embodiments herein advantageously reduce the complexity and accompanying cost of nodes in an optical network that transports WDM traffic, as compared to known networks. With reduced complexity and cost, the embodiments prove particularly useful for optically converging the traffic of multiple services. In fact, some embodiments exploit the increased traffic resulting from such convergence in order to eliminate or at least mitigate the complexity that known networks incur for flexibility in traffic routing.
More particularly, embodiments herein include a peer network node that is configured, in conjunction with other peer network nodes, to form a higher-tiered optical network that transports WDM traffic for multiple lower-tiered optical networks. These higher and lower tiered networks may respectively be a metro network and an access network, a regional network and a metro network, etc. Regardless, the peer network node comprises a plurality of dedicated bidirectional optical ports. These ports include two or more so-called lower-tiered ports and one or more so-called peer ports. Each lower-tiered port is dedicated for transporting WDM traffic to and from an individual lower-tiered network, while each peer port is dedicated for transporting WDM traffic to and from an individual peer network node. The bidirectional nature of each such ports advantageously enables WDM traffic to be transported between the peer network node and any given lower-tiered network or peer network node via a single optical fiber.
The peer network node further includes one or more so-called hub-side bidirectional optical ports. Each hub-side port is configured to transport WDM traffic to and from a hub node in the higher-tiered network. Further, at least one of the ports is a common port configured to transport WDM traffic aggregated across multiple lower-tiered ports. Similarly to the lower-tiered ports, the bidirectional nature of each hub-side port advantageously enables WDM traffic to be transported between the hub-side port and the hub node via a single optical fiber.
The peer network node finally includes a switching circuit. The switching circuit is configured to distribute WDM traffic received at the one or more hub-side ports to respective dedicated ports for dedicated transport to one or more of the lower-tiered networks and peer network nodes. Notably, the switching circuit is also configured to direct any WDM traffic received at the dedicated ports to the one or more hub-side ports for transport to the hub node, even if that traffic is actually destined for one of the lower-tiered networks to which a lower-tiered port is connected. This reduced routing flexibility, in conjunction with the bidirectional nature of the ports, advantageously reduces the complexity and cost of the peer network node, while at the same time satisfying the routing requirements for a wider range of applications.
In at least some embodiments, for example, the peer network node includes a single wavelength selective switch (WSS), which significantly reduces the complexity and cost of the node as compared to known approaches that employ a ROADM with at least two WSSs. In this and other embodiments, the node may also include a bypass path that bypasses traffic received at a peer port around any WSSs in the peer network node to a hub-side port, for transport to the hub node. Alternatively, the traffic received at the peer port may be input into a WSS, for aggregated transport to the hub with other traffic, e.g., lower-tiered network traffic.
One or more embodiments herein also provide for enhanced resiliency to faults in the network. In some embodiments, for instance, at least two different dedicated ports of the peer network node receive the same traffic. The switching circuit is configured to direct that traffic to a hub-side port by dynamically selecting from which dedicated port to acquire the traffic. This dynamic selection is based on a control signal associated with any faults in the higher or lower-tiered networks affecting those dedicated ports.
In other embodiments, the node alternatively employs redundancy for its connection to the hub node, in order to protect against faults in those connections or in any intermediate peer nodes. In this case, the node includes a redundant hub-side bidirectional optical port that is configured to transport the same traffic as another hub-side port, but to transport that traffic to and from a different, redundant hub node.
In still other embodiments, the node employs both redundancy for a lower-tiered network or a peer network node, and redundancy for its connection to the hub node. Such embodiments may utilize an additional WSS for realizing this enhanced resiliency. Some embodiments may further utilize an optical switch for protecting against a single simultaneous failure in both a connection to a lower-tiered network and a connection to a hub node.
Of course, the present invention is not limited to the above features and advantages. Indeed, those skilled in the art will recognize additional features and advantages upon reading the following detailed description, and upon viewing the accompanying drawings.
The hub node 18 connects uplink WDM traffic from one or more network nodes 16 to a higher-tiered network called the regional network. More specifically, the hub node 18 routes uplink WDM traffic to an appropriate one of multiple edge nodes (not shown), e.g., a business services edge router, a residential services or mobile services broadband network gateway (BNG), a broadband remote access server (BRAS), etc. The edge node then performs subscriber management and routes the uplink traffic (typically at the packet level) towards an appropriate destination, such as to content servicers, back towards the access networks, to the Internet, etc. Such edge node routing may entail sending the uplink traffic to the regional network, which operates back at the optical layer. Thus, although omitted from
The regional network is also formed from a plurality of interconnected peer network nodes 16, which hub WDM traffic to a hub node 18 in the regional network much the same as in the metro network. Traffic from the regional network is then placed onto a long haul network at tier 4, for inter-regional transport. Downlink WDM traffic propagates through the networks in an analogous, but opposite, manner.
Known implementations of this tiered architecture 10 configure each peer network node 16 with significant routing flexibility. Each peer network node 16, for example, includes a reconfigurable optical add/drop multiplexers (ROADM) that enables any WDM traffic to be selectively added or dropped from the node 16. Equipped with such hardware, a peer network node 16 can immediately drop any uplink traffic that is received from another peer network node 16 if that traffic is destined for a connected lower-tiered network. However, because each ROADM requires at least two wavelength selective switches (WSSs) just to provide this flexible routing capability and may require additional WSSs to provide full flexibility in adding or dropping wavelengths, known implementations prove cost-prohibitive and/or operationally limited in some contexts.
Embodiments herein advantageously reduce the complexity and accompanying cost of peer network nodes 16. With reduced complexity and cost, the embodiments prove useful in a wider range of applications, such as optically converging the traffic of multiple services. In fact, some embodiments exploit the increased traffic resulting from such convergence in order to eliminate or at least mitigate the complexity that known networks incur for flexibility in traffic routing.
The peer network node 20 further includes one or more so-called hub-side bidirectional optical ports 24. Each hub-side port 24 is configured to transport WDM traffic to and from a hub node 18 in the higher-tiered network. Further, at least one of the ports 24 is a common port configured to transport WDM traffic aggregated across multiple lower-tiered ports 22A.
Although as shown the hub-side ports of peer network node 20 transport traffic directly to and from a hub node 18, rather than indirectly via one or more other peer network nodes 16, this need not be the case. Indeed, the ports 24 are hub-side merely in the sense that traffic input into or output from such ports 24 at some point in time originates from or is destined for a hub node 18. Similarly to the lower-tiered ports 22, the bidirectional nature of each hub-side port 24 advantageously enables WDM traffic to be transported between the hub-side port 24 and the hub node 18 via a single optical fiber 14.
The peer network node 20 finally includes a switching circuit 26. The switching circuit 26 is configured to distribute WDM traffic received at the one or more hub-side ports 24 to respective dedicated ports 22A, 22B for dedicated transport to one or more of the lower-tiered networks and peer network nodes 16. Notably, the switching circuit 26 is also configured to direct any WDM traffic received at the dedicated ports 22A, 22B to the one or more hub-side ports 24 for transport to the hub node 18, even if that traffic is actually destined for one of the lower-tiered networks to which a lower-tiered port 22A is connected.
By unconditionally transporting traffic to the hub node 18 in this way, the switching circuit 26 has reduced flexibility in routing WDM traffic. Indeed, traffic received at a dedicated port 22 that is destined for a lower-tiered port 22A must first be routed to the hub node 18, and then back to the peer network node 20 before finally being routed to that lower-tiered port 22A. But as demonstrated in greater detail below, this reduced routing flexibility in conjunction with the bidirectional nature of the ports 22, 24, advantageously reduces the complexity and cost of the peer network node 20, while at the same time satisfying the routing requirements for a wider range of applications.
Consider, for example, the embodiment illustrated in
In this regard, the switching circuit 26 advantageously includes a bypass path 30. The bypass path 30 is a circuit configured to bypass traffic received at hub-side port 24B around the one or more WSSs 28 to peer port 22B, for transport to an associated peer network node 16. Likewise, the bypass path 30 bypasses traffic received at peer port 22B around the one or more WSSs 28 to hub-side port 24B, for transport to the hub node 18. The bypass path 30 transports traffic to the hub node 18 in this way, even if that traffic is actually destined for a lower-tiered network to which a lower-tiered port 22A is connected. In doing so, the bypass path 30 effectively eliminates any flexibility with which the peer network node 20 would otherwise be able to route the traffic received at ports 22B, 24B, to correspondingly reduce the complexity and cost of the node 20.
Yet even with this reduced complexity and cost, the node 20 still proves useful in a wider range of applications. In one embodiment, for example, the node 20 is configured to transport the WDM traffic of multiple services (e.g., mobile, business, and residential services) in a converged fashion. This convergence may substantially increase traffic utilization of the optical fiber 14 connecting the node 20 to a peer network node 16 at the peer port 22B, perhaps to near capacity. If so, little if any additional traffic should be added to that fiber 14 from the lower-tiered networks to which the node 20 is connected. The node 20 thereby exploits the filling of the fiber 14 to substantially full capacity as an opportunity to reduce the complexity and cost of the node, by bypassing the fiber 14 and the traffic it carries around the WSSs 28 that would otherwise add additional traffic to the fiber 14.
Of course, if any of that bypassed traffic was actually destined for one of the lower-tiered networks, the traffic may be routed back from the hub node 18 to the peer network node's hub-side port 24A. From port 24A, the traffic is then distributed to the appropriate lower-tiered network. Because in many applications this additional round-trip transport is needed only very rarely, the complexity and cost reductions obtained from the bypass path 30 still prove advantageous on balance.
Indeed, as shown in
Other embodiments herein, by contrast, contemplate that the reduction in complexity and cost results exclusively from the inclusion of a single WSS 27 in the node 20, rather than also from the inclusion of a bypass path 30.
Regardless of the particular manner in which the complexity and cost of the node 20 are reduced, that reduction permits the node 20 to realize substantially the same meaningful functionality, with less expense. The reduction may additionally or alternatively permit the node 20 to realize new functionality, such as enhanced resiliency to faults in the network.
In one embodiment, for example, the node 20 employs redundancy for a lower-tiered network or a peer network node 16 to which it is connected, in order to protect against faults in the connection to that network or node 16. In this case, at least two different dedicated ports 22 of the node 20 receive the same traffic. The switching circuit 26 is configured to direct that traffic to a hub-side port 24 by dynamically selecting from which dedicated port 22 to acquire the traffic. This dynamic selection is based on a control signal associated with any faults in the higher or lower-tiered networks affecting those dedicated ports 22.
As one example of this,
In other embodiments, the node 20 alternatively employs redundancy for its connection to the hub node 18, in order to protect against faults in those connections or in any intermediate peer nodes 16.
Of course, those skilled in the art will appreciate that
In still other embodiments, the node 20 employs both redundancy for a lower-tiered network or a peer network node 16, and redundancy for its connection to the hub node 18. Such embodiments may utilize an additional WSS for realizing this enhanced resiliency. Although this increases complexity and cost, the embodiments still achieve greater functionality than that of known approaches with comparable complexity and cost.
The switching circuit 26 includes a first WSS 40 and a second WSS 42. The first WSS 40 includes two dedicated ports that connect to or otherwise correspond to lower-tiered ports 22A-1 and 22A-2. The second WSS 42 includes two dedicated ports that connect to or otherwise correspond to lower-tiered ports 22A-3 and 22A-4. Configured in this way, the first WSS 40 is configured to direct traffic from lower-tiered network 36, as received at lower-tiered port 22A-1, to hub-side port 24A, while the second WSS 42 is configured to direct traffic from that lower-tiered network 36, as received at lower-tiered port 22A-3, to redundant hub-side port 24C. Similarly, the first WSS 40 is configured to direct traffic from lower-tiered network 38, as received at lower-tiered port 22A-2, to hub-side port 24A, while the second WSS 42 is configured to direct traffic from that lower-tiered network 38, as received at lower-tiered port 22A-4, to redundant hub-side port 24C.
While the example in
Nonetheless, as shown, the switching circuit 26 further comprises an optical switch 44. The optical switch 44 includes a first port 44A and a second port 44B. The switch 44 is configured to dynamically switch the traffic from lower-tiered network 36, as received at lower-tiered port 22A-1, to either the first or second port 44A, 44B of the optical switch 44. The switch 44 is further configured to dynamically switch the traffic from lower-tiered network 36, as received at lower-tiered port 22A-3, to either the first or second port 44A, 44B of the optical switch 44. Such dynamic switching is performed responsive to a control signal associated with any faults in the higher or lower-tiered networks affecting those ports 22A-1, 22A-3.
With the switch 44 configured in this way, the first WSS 40 is configured to direct the traffic received from the first port 44A of the optical switch 44 to hub-side port 24A. Meanwhile, the second WSS 42 is configured to direct the traffic received from the second port 44B of the optical switch 44 to redundant hub-side port 24C.
Such provides protection against a simultaneous failure in both a connection to lower-tiered network 36 and a connection to a hub node 18. Consider an example where both the connection to lower-tiered network 36 at port 22A-1 fails and the connection to redundant hub node 18C at port 24C fails. In this case, the optical switch 44 is configured to dynamically switch the traffic from lower-tiered network 36, as received at port 22A-3, to the first port 44A of the optical switch 44. The first WSS 40 correspondingly directs this traffic to the hub-side port 24A, for transport to hub node 18A.
Those skilled in the art will of course appreciate that the redundant embodiments shown in
Those skilled in the art will further appreciate that while embodiments herein reduce the routing flexibility of a peer network node 16, the embodiments actually provide more flexibility in terms of the wavelengths used to transport traffic. In this regard, the switching circuit 26 in one or more embodiments is configured to dynamically adapt the wavelengths used for transporting traffic to or from any given dedicated port 22, responsive to a control signal received from a hub node 18. This control signal may be sent, for instance, upon the introduction of additional traffic to the network, whereupon the switching circuit 26 dynamically allocates an additional wavelength for the traffic. Contrary to known ROADM architectures, dynamic allocation may entail re-assigning wavelengths across different dedicated ports 22, as needed.
Turning now to additional details of a hub node 18,
In one or more other embodiments, the wavelength controller 52 is additionally or alternatively configured to generate and send a control signal that directs a peer network node 16 to dynamically adapt the wavelengths used for transporting traffic to or from any given dedicated port 22 of that peer network node 16. Such generation may be performed responsive to determining that new traffic is to be introduced or otherwise transported by the network.
In view of the above modifications and variations, those skilled in the art will appreciate that a peer network node 16 herein generally performs the processing shown in
Likewise, those skilled in the art will appreciate that a hub node 18 herein generally performs the processing shown in
Still further, those skilled in the art will understand that no particular type of WDM is required to practice the above embodiments. Thus, the embodiments may employ coarse WDM or dense WDM. The embodiments may even be used in the context of a WDM passive optical network (WDM-PON), with or without inverse return to zero/return to zero (IRZ/RZ) wavelength re-use. In one embodiment, for instance, the embodiments utilize 25 GHz channel spacing in both C and L bands, allowing for up to 400 wavelength channels per fiber. Another embodiment utilizes up to 100 GHz channel spacing.
Likewise, no particular type of technology is required to implement the one or more WSSs employed by the above embodiments. Indeed, WSSs herein may be realized using array waveguide gratings (AWGs), microelectromechnical systems (MEMs), liquid crystal on silicon (LCoS), or any other technology that may permit selective switching of optical signals on a per-wavelength basis.
Thus, those skilled in the art will recognize that the present invention may be carried out in other ways than those specifically set forth herein without departing from essential characteristics of the invention. The present embodiments are thus to be considered in all respects as illustrative and not restrictive, and all changes coming within the meaning and equivalency range of the appended claims are intended to be embraced therein.