OTN standards pave way for next-generation optical transport


By Robert Pease

Telecommunications providers are surging ahead to meet the urgent need to manage ultra-high-capacity networks and provide the capability for evolving to a multiservice transport platform that supports packet-based data transport as well as legacy traffic. According to the International Telecommunication Union (ITU), new Optical Transport Network (OTN) standards will enable the intelligent optical networking necessary for such a transition.

The ITU-Telecommunication Standardization Sector (ITU-T) initiated work on OTN standards in 1997 with G.872, which ad dresses the OTN architecture. The recommendation was formally approved in February 1999. Last February, optical data networking entered a new era with the approval of G.709, which specifies the interfaces for interconnection between service providers and network operators and facilitates mid-span meet between equipment from different vendors.

The key to G.709 is its use of digital wrapper technology. Introduced by Lucent Technologies (Murray Hill, NJ) in 1999, digital wrapper technology provides a method for encapsulating data in a frame, or optical data unit, similar to SONET framing.

In essence, digital wrapping involves grouping a number of existing frames together into one entity that can be more efficiently managed with a small allocation of overhead and forward error correction (FEC) bytes. There are three levels of hierarchy for the encapsulation-the optical channel payload unit (OPUk) for encapsulation of client data; optical channel data unit (ODUk) for encapsulating the OPUk data; and optical channel transport unit (OTUk) for encapsulating the ODUk data. The ODUk is the electrical representation of data in which digital wrapper signals are actually transported.

The G.709 standard provides a method for adding management and intelligence to optical carriers, specifically wavelengths in DWDM systems. This methodology involves wrapping client information of any protocol in a frame carrying information about both the client and the optical carrier. The information can then be used to manage the optical signals.

"G.709 OTN adds OAM&P [operations, administration, maintenance, and provisioning] functionality to optical carriers, specifically DWDM system wavelengths," says Dr. Near Margalit, chief technical officer at Centerpoint Broadband Technologies (San Jose, CA). "This involves wrapping native client information that can be of any protocol in a frame that provides performance monitoring, fault isolation, and band management."

Currently, the digital wrapper is defined for 2.5-, 10-, and 40-Gbit/sec systems. (For more information on how digital wrapping works within G.709, see Lightwave, March 2001, page 292). Despite the current reliance on standard SONET/SDH transport, proponents of G.709 believe that acceptance and adoption of OTN is inevitable once its advantages-particularly for data transmission-become better known.

"OTN includes all the benefits of SONET performance management as fault isolation with special 'hooks' and benefits specific for multiwavelength systems," says Margalit. "In addition, the hooks in G.709 allow a signal to be managed across multiple carriers, such as in wholesale wavelength services. Each carrier can maintain its own SLAs [service-level agreements] and fault isolation with a client crossing many carrier networks."

The benefits are twofold. First, the single, unified management scheme is protocol-agnostic, allowing carriers to manage native data services. Sec ond, it enables performance management and fault isolation across multiple domains. That is critical to wholesaling wavelength services, because it allows each carrier to maintain its own SLAs and fault isolation over several carrier networks.

The developers of OTN had ambitious goals. "OTN was defined as a new transport technology targeted to support the emergence of higher-bandwidth applications and various data and voice services," says Devesh Khare, product marketing engineer for Vitesse Semiconductor Corp. (Santa Clara, CA). "It is targeted as a true replacement for SONET, although both can be used in unison."

Khare believes the technology is still too young and carriers will continue to resist giving up their SONET/SDH infrastructures for the likes of digital wrapper. Other vendors agree that an evolutionary process is a much more likely scenario for industry-wide OTN acceptance.

For example, Applied Micro Circuits Corp. (AMCC-San Diego) promotes digital wrapper as a technology that "provides an extension of the current SONET/SDH infrastructure, while enhancing both flexibility and transparency across the network," according to a statement provided to Lightwave. Using OTN in tandem with SONET/SDH may provide the best of both worlds, while technologies like digital wrapper go through the acceptance phase. That would require new DWDM systems to support management through both SONET/SDH and OTN. Allow ing carriers to mix and match OTN and SONET/SDH freely could be critical to industry acceptance as it applies to different service deployments.

"Open aspects of G.709 still need to be resolved to enable OTN deployment, including such items as electrical multiplexing and asynchronous client mapping as well as clocking and jitter requirements," says Steve Rood Goldman, product-line manager for FEC products at MultiLink Technolo gy Corp. (Somerset, NJ). "In addition, G.709 is a low-level standard describing frame structure and management fields. The standards related to the optical control plane, such as network management and connection setup and teardown, still have not been finalized."

Vendors are confident that OTN initiatives are the future of optical transport, and the movement will only grow as more benefits are realized and more G.709-compliant products are developed. "All our customers are interested in providing G.709-compliant products," says AMCC's director of marketing, Neal Neslusan. "I would say it's not a matter of 'if,' but a matter of 'when' for the carriers to implement G.709."

More in Packet Transport