{"683062":{"#nid":"683062","#data":{"type":"news","title":"Lighting the Way to Faster Data Transfer","body":[{"value":"\u003Cp\u003EThe future of computing is lit, literally.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EAs microchips grow more complex and data demands intensify, traditional electrical connections are hitting their limits. Speed is king in today\u2019s digital systems, but a major bottleneck remains in how quickly information can move between components like processors and memory.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EThis lag is one of the most pressing challenges in advanced hardware design. While processors continue to accelerate, the links that connect them can\u0027t keep pace.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EGeorgia Tech researcher \u003Ca href=\u0022https:\/\/ece.gatech.edu\/directory\/ali-adibi\u0022\u003E\u003Cstrong\u003EAli Adibi\u003C\/strong\u003E\u003C\/a\u003E is addressing this problem with $5.3 million in funding over three years from the Defense Advanced Research Projects Agency (DARPA). His project is part of DARPA\u2019s \u003Ca href=\u0022https:\/\/www.darpa.mil\/research\/programs\/happi-heterogeneous\u0022 rel=\u0022noreferrer\u0022\u003E\u003Cstrong\u003EHeterogeneous Adaptively Produced Photonic Interfaces\u003C\/strong\u003E\u003C\/a\u003E (HAPPI) program, which aims to dramatically boost the speed and density of data transmission within microsystems by using light instead of electricity.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003E\u201cOptical solutions are highly advantageous for providing the required data rates and power consumptions, and our project is formed to address the most important challenges for achieving the system-level performance,\u201d said Adibi, a professor and Joseph M. Pettit Chair in the \u003Ca href=\u0022https:\/\/ece.gatech.edu\/\u0022\u003E\u003Cstrong\u003ESchool of Electrical and Computer Engineering\u003C\/strong\u003E\u003C\/a\u003E.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EThe project brings together a multidisciplinary team, including collaborators from the Massachusetts Institute of Technology, University of Florida, NY CREATES, and NHanced Semiconductors, Inc.\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EGoing Vertical\u003C\/strong\u003E\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EUnlike traditional optical communication, which connects systems across distances, this project focuses on enabling ultra-fast, low-loss communication \u003Cem\u003Ewithin\u003C\/em\u003Eelectronic systems.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EThe key innovation is vertically connecting electronic chips in a compact stack. This design helps overcome the limitations of planar optical routing geometries (layouts that guide light horizontally across a chip) which are often not compatible with the dense, 3D chip architectures needed for next-generation computing.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EAdibi\u2019s team is developing a novel 3D optical routing system that can transmit data with minimal loss, high bandwidth, and compact components. The system is designed to scale to large arrays of interconnected chips with minimal interference between data channels.\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003ESmarter Design with Machine Learning\u003C\/strong\u003E\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EAt the heart of the project is the use of machine learning (ML) to help design and optimize the light-based communication system.\u0026nbsp;\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EML is used to shape and fine-tune the tiny structures that guide light through and between chips. This includes finding the best sizes, shapes, and layouts for components like couplers and waveguides, so they can be made smaller, work more efficiently, and fit into dense chip layouts.\u0026nbsp;\u0026nbsp;\u003C\/p\u003E\u003Cp\u003E\u201cDesigning a complete, scalable 3D optical routing structure involves innumerable variables,\u201d Adibi said. \u201cMachine learning helps us navigate that complexity and find solutions that would be nearly impossible to identify manually.\u201d\u0026nbsp;\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003ETiny \u0022Mirrors\u0022\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003EAnother key innovation involves specialized optical structures, or what Adibi refers to as \u201cartificial mirrors\u201d.\u003C\/p\u003E\u003Cp\u003EThe tiny, precisely shaped structures, called metagratings, are embedded in the chip material to redirect light vertically between layers with minimal loss. These components are designed to guide light efficiently in tight spaces, helping connect stacked chips without losing signal strength.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003E\u201cImagine light traveling through a chip and suddenly being redirected straight up. That\u2019s the kind of precise control we\u2019re achieving,\u201d Adibi explained.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EThese innovations, along with advanced techniques for building vertical light paths through thick silicon layers and new packaging solutions that keep components precisely aligned, have shown promise on their own. But combining them is what enables dense, high-speed, low-loss communication between vertically stacked chips, something that no system has achieved before, according to Adibi.\u0026nbsp;\u003C\/p\u003E\u003Cp\u003E\u201cAs with any complex system, success depends on how well everything is structured and optimized,\u201d he said. \u201cOnce everything is in alignment, data can move faster, more efficiently, and with less energy consumption for communicating each bit of data.\u201d\u003C\/p\u003E\u003Cp\u003E\u003Cbr\u003E\u003Cem\u003E\u003Cstrong\u003EAbout the Research\u003C\/strong\u003E\u003C\/em\u003E\u003Cbr\u003E\u003Cem\u003EThis research is supported by the Defense Advanced Research Projects Agency (DARPA) \u003C\/em\u003E\u003Ca href=\u0022https:\/\/www.darpa.mil\/research\/programs\/happi-heterogeneous\u0022 rel=\u0022noreferrer\u0022\u003E\u003Cem\u003E\u003Cstrong\u003EHeterogeneous Adaptively Produced Photonic Interfaces (HAPPI) program\u003C\/strong\u003E\u003C\/em\u003E\u003C\/a\u003E\u003Cem\u003E. Notice ID DARPA-SN-24-105.\u003C\/em\u003E\u003C\/p\u003E","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EDARPA is backing Professor Ali Adibi\u2019s work to use light, not electricity, to move data faster and more efficiently in next-generation electronics.\u0026nbsp;\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":[{"value":"DARPA is backing Professor Ali Adibi\u2019s work to use light, not electricity, to move data faster and more efficiently in next-generation electronics. "}],"uid":"36172","created_gmt":"2025-07-09 18:43:36","changed_gmt":"2025-07-09 18:49:29","author":"dwatson71","boilerplate_text":"","field_publication":"","field_article_url":"","location":"Atlanta, GA","dateline":{"date":"2025-07-09T00:00:00-04:00","iso_date":"2025-07-09T00:00:00-04:00","tz":"America\/New_York"},"extras":[],"hg_media":{"677375":{"id":"677375","type":"image","title":"25-2304-Darpa-Happi-Ali-Adibi-007.JPG","body":"\u003Cp\u003ESilicon-on-insulator (SOI) wafer used in a multi-chip module featuring 3D optical interconnects. \u003Cem\u003E(Photo: Allison Carter)\u003C\/em\u003E\u003C\/p\u003E","created":"1752086638","gmt_created":"2025-07-09 18:43:58","changed":"1752086638","gmt_changed":"2025-07-09 18:43:58","alt":"Photo of Silicon-on-insulator (SOI) wafer","file":{"fid":"261269","name":"25-2304-Darpa-Happi-Ali-Adibi-007.JPG","image_path":"\/sites\/default\/files\/2025\/07\/09\/25-2304-Darpa-Happi-Ali-Adibi-007.JPG","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/2025\/07\/09\/25-2304-Darpa-Happi-Ali-Adibi-007.JPG","mime":"image\/jpeg","size":1306660,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/2025\/07\/09\/25-2304-Darpa-Happi-Ali-Adibi-007.JPG?itok=b2dPJ-8H"}},"677376":{"id":"677376","type":"image","title":"MulitChip.jpg","body":"\u003Cp\u003EA schematic illustration of a multi-chip structure with 3D optical routing. The key parts of Adibi\u0027s proposed system are: 1) multi-layer planar waveguides, 2) free-form couplers, and 3) a dense vertical waveguide array.\u003C\/p\u003E","created":"1752086638","gmt_created":"2025-07-09 18:43:58","changed":"1752086638","gmt_changed":"2025-07-09 18:43:58","alt":"A schematic illustration of a multi-chip structure with 3D optical routing.","file":{"fid":"261270","name":"MulitChip.jpg","image_path":"\/sites\/default\/files\/2025\/07\/09\/MulitChip.jpg","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/2025\/07\/09\/MulitChip.jpg","mime":"image\/jpeg","size":7987738,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/2025\/07\/09\/MulitChip.jpg?itok=KD-djsVC"}},"677374":{"id":"677374","type":"image","title":"25-2304-Darpa-Happi-Ali-Adibi-006.JPG","body":"\u003Cdiv\u003E\u003Cdiv\u003E\u003Cdiv\u003E\u003Cp\u003EBy combining advanced optical techniques, Professor Ali Adibi\u2019s 3D optical routing systems looks to enable vertical chip integration in a way not previously achieved. \u003Cem\u003E(Photo: Allison Carter)\u003C\/em\u003E\u003C\/p\u003E\u003C\/div\u003E\u003C\/div\u003E\u003C\/div\u003E","created":"1752086638","gmt_created":"2025-07-09 18:43:58","changed":"1752086638","gmt_changed":"2025-07-09 18:43:58","alt":"Professor Ali Adibi in front of testing equipment for his 3D optical routing system.","file":{"fid":"261268","name":"25-2304-Darpa-Happi-Ali-Adibi-006.JPG","image_path":"\/sites\/default\/files\/2025\/07\/09\/25-2304-Darpa-Happi-Ali-Adibi-006.JPG","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/2025\/07\/09\/25-2304-Darpa-Happi-Ali-Adibi-006.JPG","mime":"image\/jpeg","size":1563309,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/2025\/07\/09\/25-2304-Darpa-Happi-Ali-Adibi-006.JPG?itok=NoOrAjDb"}}},"media_ids":["677375","677376","677374"],"groups":[{"id":"1188","name":"Research Horizons"}],"categories":[{"id":"145","name":"Engineering"},{"id":"149","name":"Nanotechnology and Nanoscience"},{"id":"194610","name":"National Interests\/National Security"},{"id":"135","name":"Research"}],"keywords":[{"id":"187915","name":"go-researchnews"}],"core_research_areas":[{"id":"39451","name":"Electronics and Nanotechnology"},{"id":"193652","name":"Matter and Systems"}],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003EDan Watson\u003C\/p\u003E","format":"limited_html"}],"email":["dwatson@ece.gatech.edu"],"slides":[],"orientation":[],"userdata":""}}}