Contributing Industry Expert		Contributing Analyst
	Tom Goodwin Vice President, Marketing and Communications, Optics Division Alcatel-Lucent		Lorenza Brescia Principal Analyst, Managing Partner Telecom Strategy Partners

Interview Summary
Tom Goodwin, vice president of Marketing and Communications for Alcatel-Lucent's Optics activities, is interviewed by Lorenza Brescia on the future role of ROADM, OTN, and cross-connects in the network, with particular attention to OPEX-reduction opportunities; 40G and 100G best alternatives for modulation; standardization and adoption; and the role that vendors, and in particular Alcatel-Lucent, will have to play to differentiate themselves in the market.

Mr. Goodwin also participated in the IEC Communications Forums at NXTcomm session The Next Step in the Packet Transport Evolution: The Virtualization of the Optical Layer toward Unified Packet Network Architectures. Visit the IEC Web site for more information on other upcoming IEC events.

Brescia: What is your opinion concerning the future role of ROADM, of OTN, and of cross-connects in the network now that we are evolving towards packet optical optimization? Is there a real improvement in OPEX to be gained by their deployment going forward, and how much implementation will they see?

Goodwin: ROADM is just the early stage of future photonic networks, and this technology is providing the first steps for OPEX advantage. Moving to full tunable ROADM architecture and automatic reconfigurability, meaning no manual intervention in field, with a distributed control plane will provide a substantial advantage for infrastructure maintenance.

With traffic growing randomly such as packet-based traffic, simulations and lab evaluations conclude that a full tunable architecture enables four times better utilization of the deployed infrastructure than the traditional ROADM with fixed add/drop. This transforms [into] OPEX saving as well as CAPEX savings.

Brescia: For 40G transport, there have been only a couple of modulation formats and minor variants in use, such as DQPSK and DPSK. But now other alternatives are being explored, such as DPQPSK and coherent receiver technology — perhaps with an eye toward gaining validation for the technology at 40G in an attempt to gain lead to market at 100G. What is your opinion of which technologies are cost optimized for 40G, how long the 40G opportunity will last, and how is Alcatel-Lucent positioning itself for the transition?

Goodwin: The right modulation format for 40G transmission varies on a case-by-case basis depending on the parameter operators want to optimize such as spectral efficiency, long-distance transmission, PMD tolerance, and nonlinear effects robustness.

DPSK has proven to be effective for longer reach in 100GHz spacing, for compatibility with existing traffic at 10G in the same line and cost-efficiency. A variant of DPSK can also fit 50GHz with minor transmission impairments.

DQPSK is instead optimizing high-PMD regional distances in 50GHz spacing, but with higher cost. PM-QPSK has the spectral efficiency to fit 50GHz grid, while coherent detection improves the ability to compensate high-PMD links and chromatic dispersion. However, it suffers very much from nonlinear impairments when it is transmitted in the same line with 10G traffic, which makes the upgrade of existing 10G links almost impossible. Furthermore, this format exploits 10G optical technology, so that the optical cost structure won't get any benefit out of 40G volume for cost-reduction purposes. It seats in between DPSK and DQPSK from a cost-structure standpoint.

DPSK and DQPSK are the new optical technologies that will boost 100G cost-efficiency and are the ones poised to get benefit out of 40G volumes. Having 100G tomorrow implies that we have to invest in a real 40G format today, as 100G will be exploiting the coherent detection as well. Alcatel-Lucent is looking at optimizing the different customers' infrastructures and requirements in the most cost-effective way while providing a credible development path toward [making] the 100G commercially deployable.

Brescia: Which are the main drivers for operators to deploy 100G systems today with respect to economic factors and technology choices? How is [it impacting] the issue of standardizations for 100G systems, and how is it having impact on price and availability? Which is in your opinion the most profitable evolution path for the service provider from 40G to 100G?

Goodwin: 100G will offer better opportunity to optimize the transport of the 10G traffic (10x10G) and create more benefit than 4x10G, even if the 100G native router interface, like it is the case for the 40G, will drive the acceleration of the development. The pending decisions of the standardization of a 100G frame format will just impact the native transport and not the multiplexing, which will take over as soon as the optical solution proves to be economically viable, thanks to 40G deployments.

Brescia: Bandwidth has always tended toward commoditization over time, and in recent years network infrastructure has been increasingly commoditized and software has played an increasingly greater role in the value that vendors bring to the table. How do you see these trends continuing or changing in the future, and do they provide opportunities for a vendor such as Alcatel-Lucent to differentiate themselves in the market?

Goodwin: Bandwidth commoditization requires a technology which has proven to be the best to reduce the cost per transmitted bit, namely WDM technology. Leveraging its innovation leadership, as well as its leadership in the WDM segment, Alcatel-Lucent has the best assets to build on top of this technology to support packet optical evolution. WDM is the recognized transmission layer. Intelligence brought into this layer will make it even more suitable. Unified control plane with packet processing, OTN, and cross-connect capabilities are the way toward evolution of modern transport solutions.