Edwin Jan Klein - Universiteit Twente

Chapter 1
Today, however, the cost of optical fiber and transceivers is a fraction of they used to
be and P2P is beginning to show several advantages over PON.
By nature, the OLT in a PON is an expensive component because it needs to be able
to send a broadband optical signal up to 20 or 30 km with enough power to light up 32
or more ONUs. Therefore an OLT doesn’t start to become economical until a large
percentage of the total supported ONUs is served [11]. In addition, because many
users share the same fiber, the bandwidth for each user will decrease when new users
are added. Furthermore, the fact that many users share the same fiber also means any
upgrade to a higher aggregate bandwidth always affects multiple users (all users will
have to upgrade simultaneously), thus becoming a costly exercise.
In contrast the P2P network can be upgraded on a per-user basis and can achieve
much higher bit rates at longer distances because no synchronization is required and
no bandwidth is shared between users.
1.3.3 Wavelength Division Multiplexed PON
Combining many of the benefits of P2P and traditional passive optical networks
Wavelength Division Multiplexed (WDM)-PON is seen as a possible successor [12].
WDM-PON uses the same passive network topology as PON but the addition of many
wavelengths implicates that many different services can be run side by side.
Customers, for instance, that require higher bit rates can then do so by transferring to
a different wavelength with this service. Changes are then only required at the head
office and the customers’ premises, keeping the PON system out in the field
unchanged.
For now, however, the biggest problem for WDM-PON is cost. For true WDM
narrowband tunable or wavelength specific lasers are required in the ONUs on the
customer side. Such ONUs, however, will be prohibitively expensive in an industry
that looks for €100 to €250 solutions.
1.4 Projects
The work presented in this thesis was carried out within the final two years of the
three year NAIS project and the first two years of the ongoing Broadband Photonics
project. An important goal in both these projects was to create cost-effective devices
for use in WDM-PON networks.
1.4.1 NAIS
The EC funded NAIS (Next generation Active Integrated optic Sub-systems) project
[13] aimed to exploit recent insights in electro-optical materials and technological
advances in optical Microring Resonators to create a densely integrated optic
subsystem (ONU) with an application in WDM-PON. The technological scope of the
subsystem, shown in Figure 1.5, sees functions such as optical switching, modulation,
multiplexing and filtering integrated on a single optical chip. Only a high-speed
detector and a broadband light source, which through the use of integrated optical
filters can be implemented using a low-cost LED, are kept off-chip.
6