4 November 2004 Planar integrated free-space optics for optical interconnects and fan-out/in operations
Author Affiliations +
The advantages of optical interconnections, like low latency and large bandwidth, are currently becoming more important for on-board or multi-board systems. Clock distribution and data transfer with high channel density and high frequency over areas in order of several centimeters up to a few meters are of interest, e.g., in multi-processor systems. As an approach to solve the challenges of interconnections and fanning we investigate planar integrated free-space optical systems (PIFSO). In the "High-speed Opto-eLectronic Memory System" project (short: HOLMS), supported by the European Commission, a multi-processor system with optical processor memory interconnection is being demonstrated. The system combines different optics technologies. The PIFSO technology is used as the interface between opto-electronic components and optical fibers and/or PCB-embedded waveguides. The tasks of this interface are to realize the pitch transfer, e.g. from the waveguide array to the opto-electronic device array, fan-in and fan-out. To realize the different aspects of these tasks, suitable micro-optical imaging techniques and fan-out approaches are considered. To handle the large numerical aperture of the optical multimode field emerging from the VCSEL-diodes and the waveguides (up to around 0.2), the concept of pupil division is being investigated for the fan-out operation. Practical aspects such as footprint and tolerances are also considered. The feasibility of implementation of the PIFSO interconnect will be shown by means of simulations and experimental demonstration.
© (2004) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Manfred Jarczynski, Manfred Jarczynski, Juergen Jahns, Juergen Jahns, "Planar integrated free-space optics for optical interconnects and fan-out/in operations", Proc. SPIE 5556, Photonic Devices and Algorithms for Computing VI, (4 November 2004); doi: 10.1117/12.557175; https://doi.org/10.1117/12.557175


Back to Top