Distributed AI/ML at the Resource-Constrained Edge
Co-located with MILCOM 2025
Program
Los Angeles, CA, USA, Oct. 6-10 2025
Distributed AI/ML at the Resource-Constrained Edge
Co-located with MILCOM 2025
Program
Los Angeles, CA, USA, Oct. 6-10 2025
0800 - Welcome & Overview of Workshop
0810 - Keynote Talk: Prof. David Love, Purdue
Title: Distributed Learning: General Results and Applications to “Unsolved” Communications Problems
Abstract: The projected growth in demand for wireless broadband applications is driving industry and academic research. To enable future systems, it is critical to look beyond the waveforms and signal processing schemes of today. One of the most talked about 6G innovations is using data-driven learning techniques for network optimization and operation. In this talk, I will first start out by discussing some of the challenges with distributed learning. I will then look at the impact of both AI/ML and new applications on physical layer design. I will review the use of channel output feedback in additive white Gaussian channels and discuss the surprising (and sensitive) classical results on capacity-achieving linear processing schemes. The practical shortcomings of these kinds of linear schemes will be overviewed. I will show how an AI/ML approach can outperform the best-known schemes.
Bio: David J. Love is the Nick Trbovich Professor of Electrical and Computer Engineering at Purdue University. He received the B.S., M.S.E., and Ph.D. degrees from UT-Austin. He is currently a Senior Editor for IEEE Journal on Selected Areas in Communications and held editorial positions for IEEE Signal Processing Magazine, IEEE Trans. Communications, and IEEE Trans. Signal Processing. He holds 32 U.S. patents. His research interests include 6G and beyond wireless, MIMO communications, millimeter-wave wireless, software-defined radios, and coding theory. He is a Fellow of the IEEE, American Association for the Advancement of Science (AAAS), and National Academy of Inventors (NAI). His research has been recognized by the IEEE Communications Society (2016 Stephen O. Rice Prize, 2020 Fred W. Ellersick Prize, and 2024 William R. Bennett Prize), IEEE Signal Processing Society (2015 SPS Best Paper Award), and IEEE VT Society (2010 Jack Neubauer Memorial Award).
0910 - Session # 1:
25 min - MAPPO for Edge Server Monitoring. Samuel G Chamoun, Christian McDowell and Robin M Buchanan (Auburn University, USA); Kevin S Chan (DEVCOM Army Research Laboratory, USA); Eric Graves (Army Research Lab, USA); Yin Sun (Auburn University, USA)
25 min - VF-RRT*-PSO: A Decentralized Planning Framework for Multi-UAV Systems in Cluttered Environments. Mohammad Ibrahim (Fraunhofer FKIE, Germany & Universtität Koblenz, Germany); Thies Möhlenhof (Thales Deutschland, Germany); Andreas U. Mauthe (University of Koblenz, Germany)
1000 - Break
1020 - Session #2:
25 min - FUSCO: Distributed Agentic Data Acquisition and Fusion for Semantic Corner Case Detection. Malik Luti (University of California Irvine, USA); Marco Levorato (University of California, Irvine, USA); Juraj Gazda (Technical University of Kosice, Slovakia); Ian Andrew Harshbarger (University of California, Irvine, USA); Matúš Dopiriak (Technical University of Košice, Slovakia); Hari Pathanjaly (University of California Irvine, USA)
25 min - STAC: Spatio-Temporal Associations for Cross-Camera Analytics in Bandwidth-Constrained Environments. Ragini Gupta (University of Illinois at Urbana-Champaign, USA); Lingzhi Zhao (University of Illinois Urbana-Champaign, USA); Jiaxi Li (University of Illinois at Urbana-Champaign, USA); Volodymyr Vakhniuk (UIUC, USA); Claudiu Danilov (Boeing Research & Technology, USA); Josh Eckhardt (Boeing, USA); Keyshla Bernard (Boeing Research and Technology, USA); Klara Nahrstedt (University of Illinois Urbana-Champaign, USA)
1110 - Session # 3:
25 min - Blockchain-Enabled Distributed Intrusion Detection System for Securing IoT Networks. Charles Stolz and Jielun Zhang (University of North Dakota, USA)
25 min - Passband Signal Detection at the Edge. Gavin Parpart, Peter Martin, Stephen Jones, Noah Elmore and Jeremiah Rounds (Pacific Northwest National Laboratory, USA)
1200 - Closing Workshop Remarks & Workshop Adjourns