FOE 2009: Industry Requires Further Cost Reduction

By: By Takehito Ichida, Analyst, Ovum
( 1 May 2009 )

The ninth Fiber Optics Expo (FOE), the biggest optical fiber communication event in Asia, took place in Tokyo on 21–23 January, with a total of 15,499 registrants. Although uncertainties about the future of the industry hung over the show, there was one certainty: that further attempts at cost reduction are required.

One positive announcement came at the keynote session, where Noritaka Uji, Representative Director and Senior Executive Vice President of NTT, committed to the planned continuous investment in FTTH development. However, he also pointed out that the company would need to carefully watch and assess the impact from the current financial turmoil.

Sean Dolan, President, Asia Pacific, Alcatel-Lucent, quoted Charles Darwin: “It is not the strongest of the species that survive, nor the most intelligent, but the one most responsive to change”. Given the current situation, this message carried conviction. The more critical a problem is, the more people overtax their brains to fix it; however, people may need to adapt in order to survive in a new environment. The bottom line is, we should watch how the environment is changing carefully and respond to it swiftly. However, as Uji mentioned, in order to realize this costs will have to decrease further. In general this will push systems vendors and then component suppliers to squeeze the margins.

MANY TECHNOLOGY OPTIONS, BUT COST IS THE DRIVER
Unlike the major Japanese carriers, other Asian carriers are currently considering both GE-PON and GPON. Although Chunghwa Telecom (CHT) has deployed GE-PON since 2007, the Taiwan service provider is currently testing GPON. Indian service provider Bharat Sanchar Nigam has selected GPON for higher density city areas and GE-PON for less dense areas. It does make economic sense for service providers to be considering both available PON technologies at the same time, as using both will enable them to make efficient use of bandwidth. However, we believe that if a vendor tries to develop and market both GE-PON and GPON the investment will be too costly.

The overall understanding among the industry it that Japan is likely to start deploying 10GE-PON soon after standardization is completed. Naoto Yoshimoto of NTT Access Service System Labs discussed both the progress and challenges of 10GE-PON technology development. Although at R&D the technologies seem mature, the issue that he raised as a conclusion was, again, the cost of components – specifically 10G-uncooled LD as well as 10G APD.

Cost was also the key issue to arise from the discussion about 100G Ethernet development. Osamu Ishida of NTT Network Innovation Labs pointed out that the technology initially developed for enterprise networks is dominating telecoms networks today, and he posed the question as to who should secure the cost of investment for future technologies. For IEEE802.3ba, the development cost of serializer/deserializer (SERDES), an electric component, is overwhelmingly high. We believe that one solution may be for telecoms and datacoms to collaborate and establish a non-profit R&D center to develop such basic technologies as SERDES.

Ironically, some of the most positive feedback came from the most costly segment – undersea cable. Yasuhiro Aoki, Submarine Network division, NEC, told us that demand had remained strong even after the financial turmoil surfaced. He believes that demand is unlikely to decrease significantly in the near future. No projects were cancelled in 2008, although some projects may be delayed in the future. In general, undersea cable projects cannot stop midway as the duration is much longer than terrestrial builds and built-to-order equipment requires sponsors (service providers or a consortium of service providers) to pay up-front for each phase of the job.

Fujitsu demonstrated 1×9 wavelength selective switch (WSS) that the company developed with MEMS available in the market. The WSS implemented by Japanese system vendors and carriers today is developed by non-Japanese suppliers. Having the capability to produce such a key component as WSS may improve the company’s production of ROADM in the future and reduce costs.