Skip to main navigation Skip to search Skip to main content

Edge-based Computing Challenges and Opportunities for Sensor Fusion: Panel Review

  • Erik Blasch
  • , Genshe Chen
  • , Yu Chen
  • , Andreas Savakis
  • , Fred Daum
  • , Lynne Grewe
  • MOVEJ Analytics
  • Intelligent Fusion Technology, Inc.
  • Rochester Institute of Technology
  • RTX Corporation
  • California State University East Bay

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Sensor, data, and information fusion techniques are typically implemented in a centralized approach that requires cloud servers to process the large amounts of data. Recently, collaborative computing approaches can support effective and efficient distributed and decentralized information fusion communication among many sensors at the edge. The panel highlighted opportunities and challenges of edge computing sensor fusion designs. Examples of opportunities included decentralized timely response, extended coverage, and resilient redundancy; all of which will be developed in the near future to enhance healthcare, smart cities, and surveillance. The challenges to make these opportunities a reality would include low size, weight, and power (SWaP) constraints, privacy and security concerns, as well as standardized data flow and architecture protocols. Common to the discussion was that with more heterogeneous edge sensing, data and information flow techniques are needed to harness the prospects of a distributed enterprise data fusion network.

Original languageEnglish
Title of host publicationSignal Processing, Sensor/Information Fusion, and Target Recognition XXXIV
EditorsIvan Kadar, Erik P. Blasch, Lynne L. Grewe
PublisherSPIE
ISBN (Electronic)9781510687479
DOIs
StatePublished - 2025
EventSignal Processing, Sensor/Information Fusion, and Target Recognition XXXIV 2025 - Orlando, United States
Duration: Apr 14 2025Apr 16 2025

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume13479

Conference

ConferenceSignal Processing, Sensor/Information Fusion, and Target Recognition XXXIV 2025
Country/TerritoryUnited States
CityOrlando
Period04/14/2504/16/25

Keywords

  • Active Learning
  • Deep Learning
  • Edge Computing
  • Information Fusion
  • Transformers

Fingerprint

Dive into the research topics of 'Edge-based Computing Challenges and Opportunities for Sensor Fusion: Panel Review'. Together they form a unique fingerprint.

Cite this