Abstract

Optical access networks have been widely investigated and trialed, particularly those based on passive optical networks (PONS).¹ In most of these systems the fiber capacity is split between users by using a TDMA protocol.^2,3 Broader reach systems have been based on a mixture of WDM and TDMA, such as the SuperPon.⁴ Other proposed approaches have been based on optical CDMA⁵ and also on spectrally sliced sources, a simple wavelength coding scheme.^6,7,8 The few commercially launched systems have only been built using point to point optical gigabit Ethernet or based on the FSAN APON standard.⁹ The downside of the point to point system is that it needs a large number of fibers in the field and loads the central office end with space consuming and expensive interface technology. The current APON, on the other hand, can’t support bandwidths needed for future broadband services such as streaming HDTV class video, quality videoconferencing or massive data or movie downloads. APON’s drawback is also its linkage to ATM at a time when the focus in access technologies is being shifted towards Ethernet.

© 2002 Optical Society of America