Could optical computing be poised for mainstream aerospace and defense information processing?
THE MIL & AERO BLOG – Optical computing has been "the next big thing" now for decades. This enabling technology holds so much potential because it offers huge advances in data throughput, information security, and resistance to the effects of electronic warfare (EW) jamming, and immunity to electronic interference.
The key to optical computing involves using optical fiber or free-space lasers to replace copper electronic interconnects in chip-to-chip, box-to-box, and system-to-system applications. There have been advances in box-to-box and even board-to-backplane optical computing as fiber and laser interconnects become more accessible.
Slow to develop
Despite its promise, however, optical computing has been slow in coming. Industry advances in data throughput over conventional copper interconnects offers more affordable solutions with less technological risk than optical computing. There's a broad installed base of copper-based computing that offers economy of scale. Making the move to optical computing always has seemed like a bridge too far. One of the toughest nuts to crack revolves around optical chip-to-chip interconnects.
The roadblock in optical chip-to-chip interconnects may be loosening with new research projects that involve developing optical chip-to-chip interconnects, as well as quantum computing.
Related: Optical computing -the next revolution
Researchers at the U.S. Defense Advanced Research Projects Agency (DARPA) in Arlington, Va., announced three contracts in March for exploration of into 3D chip-to-chip and intra-chip photonic interconnects to speed information throughput and reduce vulnerability to electromagnetic interference.
DARPA is asking SRI International in Menlo Park, Calif.; the RTX Raytheon segment in Arlington, Va.; and North Carolina State University in Raleigh, N.C., to participate in the Heterogenous Adaptively Produced Photonic Interfaces (HAPPI) program.
Chip optical links
HAPPI focuses on high-density 3D chip optical links and the ability to provide several routing planes within a photonic integrated circuit or photonic interposer.
SRI, RTX Raytheon, and N.C. State will demonstrate low-loss, high-density optical interconnects for 3D chips using a scalable manufacturing process that is compatible with microelectronics. The project emphasizes vertical connections between routing layers that can traverse substrate thickness, and surface methods for coupling light from one photonic chip to another. Chip-to-fiber coupling and chip-to-chip edge coupling are not part of the program.
The HAPPI program aims to create a 1000x increase microsystem information transmission density by exploiting photonic signaling. The ability to move and process information efficiently throughout a microsystem requires signal routing technology with high data rates and dense access points. Interfaces should resist the effects of typical microsystem misalignments due to fabrication and assembly variability, especially for large link arrays that span reticle-or wafer-scale systems.
The HAPPI program is a 36-month, two-phase program, with a base and an option period. In this, the 18-month first phase the three research organizations seek to prove the feasibility of 3D routing in integrated photonics. The 18-month second phase will scale the density and prove the manufacturability of the 3D routing platform.
Robust adaptive interfaces
The three organizations also will pursue robust-by-design or adaptive interfaces capable of environmentally and mechanically stable optical performance, with compatibility with standard microelectronics manufacturing and assembly flows.
Contractors will take-on challenges of coupling to a photonic integrated circuit with demonstrated optoelectronic sources, amplifiers, modulators, multiplexers, filters, detectors, and other electro-optical components. The operating wavelengths may be within the visible or near infrared optical bands.
The project isn't talking big money yet; funding so far falls just short of $20 million divided among SRI, Raytheon, and NC. State. Still, it's an attempt at technology breakthroughs that have the potential to move optical computing into the mainstream.
John Keller | Editor-in-Chief
John Keller is the Editor-in-Chief, Military & Aerospace Electronics Magazine--provides extensive coverage and analysis of enabling electronics and optoelectronic technologies in military, space and commercial aviation applications. John has been a member of the Military & Aerospace Electronics staff since 1989 and chief editor since 1995.
