160 Gb/s BERT facility

- Non-linear Signal Processing

The high-speed facility enables the transmission, detection and quality evaluation of optical signals at 160 Gb/s and lower data rates. It is used to measure the impact of optical components on the quality of optical signals. Projects of optical regeneration and optical performance monitoring are being supported by this facility.

The heart of the system is an optical clock generating high quality short optical pulses at 40 GHz repetition. Data is encoded onto the clock at 40 Gb/s using an electrical pattern generator and electrical to optical switch. Four 40 Gb/s signals are optically interleaved in an optical time division multiplexing (OTDM) circuit to produce a 160 Gb/s signal.

Following transmission through the device under test, a 40 Gb/s signal is de-multiplexed from the 160 Gb/s signal and converted to an electrical signal. The Bit Error Rate Tester (BERT) measures the transmission performance of each 40 Gb/s signal. Signal quality is also evaluated using a high-bandwidth oscilloscope with electrical and optical plug-ins. The system also employs 40 GHz electrical clock generators, erbium-doped fiber amplifiers and other optical and RF components.

An upgrade of the existing 160 Gbit/s optical test bed is underway in 2007, allowing testing of components, devices and transmission in real-world environment. The upgrade will enable advanced modulation formats in a recirculating fibre loop link.

The recirculating loop configuration
The upgrade will be primarily based on a recirculating loop configuration consisting of a limited amount of optical fibre (~100km), making it possible to conduct experiments comparable to long haul (i.e. trans-pacific) installed optical fibre links within the environment of the lab. Light pulses can be made to circulate over and over again over the same fibre in this loop configuration. Many state-of-the-art research labs around the world often test novel devices and solutions on a test bed like this, as an alternative to expensive testing time on an existing, installed network.

The conception of such a system is well documented in literature, and consists of several pieces of equipment. Timing is everything in a system like this, and the project offers students an exciting mix between hands-on experimental work with state-of-the-art fibre optics, cutting edge applied physics and software engineering to orchestrate the whole system in a smooth and flexible way.

The upgrade
The upgrade will consist of three main parts: the transmitter setup, the fibre loop link and the receiver setup.

• Differential Phase Shift Keying (DPSK) is one of the more spectrally efficient modulation formats that is also suitable for long haul, high capacity transmission systems. The existing multi-channel (Wavelength Division Multiplexed) system will be upgraded with specific modulating equipment to implement this advanced modulation format.

• With suitable timing and gating equipment, a single 100 km fibre link in the lab will be converted into a fibre loop. Fast acousto-optic modulators with a high extinction ratio and optical amplifiers are needed for proper operation of this loop.

• Because of the more complex nature of the DPSK format and the very long transmission distances, the receiver side becomes more involved. An upgrade to a balanced receiver for demodulating the DPSK format is needed as well as active clock recovery, because of quickly varying fibre lengths because of small temperature variations.

An Outlook
This long-haul transmission test bed will be used extensively for various sorts of experiments like testing components developed by CUDOS, and will also be a critical part of the ongoing collaboration with our industrial partner Optium. This test bed will strengthen CUDOS' position in the high speed, high capacity optical systems research field.

All optical signal regeneration and processing are attracting much attention lately, as there is potential to increase the over-all bandwidth and transmission distance of the optical signal carried by a fibre. CUDOS is now at a stage where it is demonstrating several signal processing devices based on the chalcogenide platform. To further validate their behaviour in optical systems, access to long-haul system experiments is necessary for proper testing of the respective devices.

Together with Optium, CUDOS is looking at reconfigurable routing of optical signals through complex and agile networks. The long-haul fibre loop test bed will be a cornerstone for developing more elaborate system tests and experiments. Also, novel methods for optically demodulating the DPSK signals will be investigated with this test bed.

Furthermore, this system is very flexible and future-proof, as it can easily be adapted for different types of experiments. It is envisaged that the whole setup will quickly become a workhorse for many CUDOS projects.