OXC networking experiments – Nonlinear and Quantum Optics Lab

Ultra-long-haul DWDM network with 320 x 320 wavelength-port “broadcast and select” OXCs

This work was done in collaboration with I. Tomkos, J.-K. Rhee, M. Mehendale, B. S. Hallock, B. K. Szalabofka, M. Williams, S. Tsuda, and M. Sharma at Corning Inc.

Summary: We have performed a network experiment using 320×320 wavelength-port “Broadcast & Select” optical cross-connects (OXC). 80×10.7Gb/s NRZ transmission over 4160km of all-Raman amplified dispersion-managed fiber (52x80km spans) and 13 concatenated OXCs has been demonstrated. The impact of a large number of crosstalk terms has been investigated.

For details, see the following publication:

M. Vasilyev, I. Tomkos, M. Mehendale, J.-K. Rhee, A. Kobyakov, M. Ajgaonkar, S. Tsuda, and M. Sharma, “Transparent Ultra-Long-Haul DWDM Networks with “Broadcast & Select” OADM/OXC Architecture,” J. Lightwave Technol. 20, 2661 (2003).

Fig. 1. Network scenario modeled in our OXC experiment.

Fig. 2. Schematic of our 320×320 “broadcast and select” OXC. Each input signal is broadcast to all output fiber ports. Wavelength blockers WB select the channels that need to be coupled to each port.

Fig. 3. Experimental recirculating-loop setup. AOS – acousto-optical switches, RA – Raman amplifiers, DMF – symmetric dispersion-managed fiber.

Fig. 4. Q-factor of 80 channels, measured at various signal polarizations. 40 dB is the extinction ratio of the WB. To find the impact of crosstalk, we artificially degraded this extinction ratio to 32 dB, which resulted in a significant system penalty, but still error-free performance if potential FEC gain was taken into account.

Fig. 5. Q-factor dependence on the level of crosstalk per wavelength blocker.

The OXC networking experiment was carried out at Corning Incorporated’s Photonic Research and Test Center in Somerset, NJ.