{"669682":{"#nid":"669682","#data":{"type":"news","title":"TRIAD Streamlines Edge Processing of Data in Phased-Array Antennas","body":[{"value":"\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EAs the number of elements on phased array antennas continues to grow, so does the volume of data that must be processed to extract information from the signals gathered. Researchers at the Georgia Institute of Technology have developed a new approach to intelligently process that data closer to where it is generated - on the antenna subarrays themselves. \u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003ECombining technologies including machine learning, field-programmable gate arrays (FPGAs), graphics processing units (GPUs), and a new radio-frequency image processing algorithm, the research has streamlined the modular handling of radar signals to reduce processing time and cost. The improvements \u2013 as much as two or three orders of magnitude \u2013 could lead to real-time analysis of RF image data from sources ranging from potential enemy targets to speeding automobiles headed toward collisions.\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EThe research, which has been tested on a 16-element digital antenna array, was funded by the Defense Advanced Research Projects Agency\u2019s (DARPA) Tensors for Reprogrammable Intelligent Array Demonstrations (TRIAD). While the project has so far focused on real-time imaging operations on vast amounts of data, it supports the conventional beamforming operations also done by phased arrays.\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003E\u201cThe goal is to push processing up front, to where all the raw data is coming in,\u201d said Ryan Westafer, a principal research engineer at the Georgia Tech Research Institute (GTRI). \u201cWe work to manage the high-dimensional data there and extract features in real-time. With so many data sources from autonomous vehicles to drones, we can\u2019t be sharing all those raw data feeds. We need to be analyzing the data locally and sharing only the information content \u2013 the relevant features.\u201d\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EWith potentially hundreds or even thousands of subarrays generating terabytes of data every second, Westafer says this \u201cedge intelligence\u201d can pull out the desired information in real-time, allowing defense and transportation applications alike to get the important details right away \u2013 when they need it \u2013 without waiting for processing by backend servers.\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003E\u201cClassical approaches process the data in the analog format, choosing only certain components of the vast information flow for digitizing where needed,\u201d noted Alex Saad-Falcon, a Georgia Tech Ph.D. student and former GTRI researcher who co-led the project. Other portions of the data can be stored on a server for later analysis.\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003E\u201cWe want to digitize all of the data, then off-load a smaller digital portion to be shared,\u201d he said. \u201cThat gives more flexibility to antenna array algorithm designers, because it is much easier to create an algorithm in the digital domain because you can write it in code, versus analog, where you have to design a circuit and get it built. That also facilitates reprogramming when conditions change.\u201d\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EFPGAs and GPUs are keys to Georgia Tech\u2019s modular TRIAD approach. With low power consumption and high processing speeds, the FPGAs are located adjacent to the analog-to-digital converters on antenna subarrays. With help from graphics processing units (GPUs), they process the data, quickly sending it to a CPU where information from other subarrays is aggregated.\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EAs a key feature of the project, GTRI researchers collaborated with academic researchers in Georgia Tech\u2019s \u003Ca href=\u0022http:\/\/www.ece.gatech.edu\u0022\u003ESchool of Electrical and Computer Engineering\u003C\/a\u003E (ECE) to utilize SoloPulse, a new array processing algorithm designed for radio-frequency images generated in synthetic aperture radars (SAR). \u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003E\u201cThe algorithm provides an estimate of energy coming from different points in the vicinity of the array,\u201d Saad-Falcon explained. \u201cThat allows you to form an image, though you have some uncertainty about where the actual source is. The goal was to train the machine learning model to reduce that uncertainty, or learn from it to predict the source location.\u201d\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EThough SoloPulse was not originally designed for the purpose the GTRI researchers needed, their collaborators \u2013 ECE Professor \u003Ca href=\u0022https:\/\/ece.gatech.edu\/directory\/christopher-f-barnes\u0022\u003EChristopher Barnes\u003C\/a\u003E and Research Technologist J. Michael McKinney \u2013 supported its adaptation to the TRIAD goals.\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EProgramming in the digital domain can utilize tensors, which are multilinear algebraic entities that describe the relationships between objects in terms of scalars and vectors. Utilizing tensor operations also allows data representations to be shared with machine learning algorithms such as deep neural networks, which can learn how to improve their operation every time they receive new data.\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003E\u201cYou funnel the data into the new artificial intelligence tensor operations, which you also bundle up, and then at the end you get a detection, some kind of an end result that is human-actionable,\u201d said Saad-Falcon. \u201cThe whole idea is that because you frame both the traditional algorithms and the machine learning algorithms in the same format as these tensor operations, you can effectively chain them together and get speedups that you wouldn\u2019t be able to get otherwise.\u201d\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EBeyond accelerating the data processing, the use of FPGA and GPU chips could help conserve power, which can be critical for mobile applications. \u201cYou have a finite compute budget on the array, so you need to intelligently allocate the computation and use an algorithm that extracts the information you want from the signal most effectively,\u201d he said. \u201cThis is of interest to a lot of different applications in the industry right now.\u201d\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EPart of the project\u2019s goal was a demonstration to process radar pulses received by the 16-element array. The researchers used a moving emitter on a turntable in their lab to evaluate TRIAD\u2019s imaging ability. \u201cWe could immediately see the result and our total latency from emitter motion to screen update was on the order of about 20 milliseconds \u2013 almost faster than the human eye can see.\u201d\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EThe DARPA project concluded in December 2022 and the researchers are now looking at other potential applications for the technologies. Among the possible uses is shared perception, which could have applications in autonomous vehicle networks, both for commercial and defense needs.\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E\u003Cspan\u003EIn addition to those already mentioned, the research included Jonathan Andreasen and Clayton Kerce from GTRI, and Jonathan Beaudeau from Pareto Frontier LLC, who supported the FPGA digital signal processing (DSP) component of the project.\u003C\/span\u003E\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EWriter: John Toon (john.toon@gtri.gatech.edu)\u003Cbr \/\u003E\r\nGTRI Communications\u003Cbr \/\u003E\r\nGeorgia Tech Research Institute\u003Cbr \/\u003E\r\nAtlanta, Georgia\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe\u0026nbsp;\u003Ca href=\u0022https:\/\/gtri.gatech.edu\/\u0022\u003E\u003Cstrong\u003EGeorgia Tech Research Institute (GTRI)\u003C\/strong\u003E\u003C\/a\u003E\u0026nbsp;is the nonprofit, applied research division of the Georgia Institute of Technology (Georgia Tech).\u202fFounded in 1934 as the Engineering Experiment Station, GTRI has grown to more than 2,900 employees, supporting eight laboratories in over 20 locations around the country and performing more than $800 million of problem-solving research annually for government and industry.\u202fGTRI\u0027s renowned researchers combine science, engineering, economics, policy, and technical expertise to solve complex problems for the U.S. federal government, state, and industry.\u003C\/p\u003E\r\n","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003EAs the number of elements on phased array antennas continues to grow, so does the volume of data that must be processed to extract information from the signals gathered. Researchers at the Georgia Institute of Technology are working to develop a new approach that could lead to real-time analysis of RF image data from sources ranging from potential enemy targets to speeding automobiles headed toward collisions.\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n","format":"limited_html"}],"field_summary_sentence":[{"value":"Researchers at the Georgia Institute of Technology have developed a new approach to intelligently process data on phased array antennas, reducing processing time and cost."}],"uid":"35832","created_gmt":"2023-09-15 13:49:37","changed_gmt":"2023-09-15 13:57:58","author":"Michelle Gowdy","boilerplate_text":"","field_publication":"","field_article_url":"","dateline":{"date":"2023-09-15T00:00:00-04:00","iso_date":"2023-09-15T00:00:00-04:00","tz":"America\/New_York"},"extras":[],"hg_media":{"671715":{"id":"671715","type":"image","title":"GTRI TRIAD demonstration setup","body":"\u003Cp\u003E\u003Cem\u003EImage shows the final TRIAD demonstration setup, with the transmit antenna in the foreground on a metal arm attached to a turntable and the elemental digital array in the background. Shown are Ryan Westafer and Alex Saad-Falcon. (Credit: Sean McNeil, GTRI)\u003C\/em\u003E\u003C\/p\u003E\r\n","created":"1694784587","gmt_created":"2023-09-15 13:29:47","changed":"1694784826","gmt_changed":"2023-09-15 13:33:46","alt":"GTRI TRIAD demonstration setup","file":{"fid":"254821","name":"TRIAD-Phased-Array_06.jpg","image_path":"\/sites\/default\/files\/2023\/09\/15\/TRIAD-Phased-Array_06.jpg","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/2023\/09\/15\/TRIAD-Phased-Array_06.jpg","mime":"image\/jpeg","size":2672280,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/2023\/09\/15\/TRIAD-Phased-Array_06.jpg?itok=IVm-3w6B"}},"671716":{"id":"671716","type":"image","title":"GTRI final TRIAD demonstration setup","body":"\u003Cp\u003E\u003Cem\u003EImage shows the final TRIAD demonstration setup, with the transmit antenna in the foreground on a metal arm attached to a turntable, and the elemental digital array in the background. Shown are Ryan Westafer (left) and Alex Saad-Falcon, who is holding a metal screen to show the effect of adding an additional scatterer. \u0026nbsp;(Credit: Sean McNeil, GTRI)\u003C\/em\u003E\u003C\/p\u003E\r\n","created":"1694784864","gmt_created":"2023-09-15 13:34:24","changed":"1694784959","gmt_changed":"2023-09-15 13:35:59","alt":"GTRI final TRIAD demonstration setup","file":{"fid":"254822","name":"TRIAD-Phased-Array_03.jpg","image_path":"\/sites\/default\/files\/2023\/09\/15\/TRIAD-Phased-Array_03.jpg","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/2023\/09\/15\/TRIAD-Phased-Array_03.jpg","mime":"image\/jpeg","size":2582479,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/2023\/09\/15\/TRIAD-Phased-Array_03.jpg?itok=UKeajaXS"}}},"media_ids":["671715","671716"],"groups":[{"id":"1276","name":"Georgia Tech Research Institute (GTRI)"},{"id":"1188","name":"Research Horizons"}],"categories":[{"id":"135","name":"Research"},{"id":"147","name":"Military Technology"},{"id":"129","name":"Institute and Campus"}],"keywords":[{"id":"416","name":"GTRI"},{"id":"365","name":"Research"},{"id":"187915","name":"go-researchnews"},{"id":"166902","name":"science and technology"},{"id":"2616","name":"antenna"},{"id":"690","name":"darpa"},{"id":"2435","name":"ECE"},{"id":"175350","name":"TRIAD"},{"id":"7638","name":"phased-array"}],"core_research_areas":[{"id":"39481","name":"National Security"}],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E(Interim) Director of Communications\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003EMichelle Gowdy\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003EMichelle.Gowdy@gtri.gatech.edu\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cspan\u003E\u003Cspan\u003E404-407-8060\u003C\/span\u003E\u003C\/span\u003E\u003C\/p\u003E\r\n","format":"limited_html"}],"email":["michelle.gowdy@gtri.gatech.edu"],"slides":[],"orientation":[],"userdata":""}}}