Skip to main content
Mathematics and Computer Science Division

SAGE: A Software-Defined Sensor Network

SAGE will build a national research infrastructure of new sensors that support programmable edge computers and machine learning within an interconnected cyberinfrastructure, spanning multiple major science instruments

Project Overview

SAGE nodes will provide a set of baseline measurements across all partner instruments, along with enhanced sensors selectively installed at the partner sites. The project will leverage experience gained through the successful deployment of hundreds of commercially built and tested nodes in the AoT project (Chicago and partner cities), implementing Waggle’s resilient node design to support reliable sensing and edge computation. Building on Waggle’s end-to-end data pipeline, integrated data repository, and data APIs, SAGE will develop a common programming environment, empowering NEON, ARM, HPWREN, UWI, and AoT users to collaborate and add new sensing and analysis capabilities.

SAGE will also integrate a new generation of ML hardware into Waggle nodes, extending AoT demonstrated capabilities for complex analysis of behaviors in both natural environments (flora and fauna) and human-centered ones, and implementing multimessenger adaptive measurement capabilities.

Combined, these new capabilities will create an integrated research infrastructure to enable scientists to test existing and new theories, from the Human-Technology Frontier in Smart and Connected Communities to the Rules of Life. Moreover, SAGE will build on STEM education and outreach programs already successfully established by each of the partners. For instance, the Lane of Things (LofT) program in Chicago has trained over 450 students in measurement-based science and wireless sensor design and deployment.

Novel Contributions of SAGE

SAGE provides novel capabilities to explore complex, convergent research questions spanning natural and built environments, from neighborhood to continental scale:

  • Adaptive instrumentation that responds to local condition and events, linking observations in real time, spanning ecosystems, atmosphere, urban, and extreme events (e.g., wildfires, heatwaves, storms).
  • Research Instrumentation to improve and synergize sensing efforts across NSF, with new flexible, interconnected resources to support research integrating BIO, CISE, ENG, GEO, and SBE.
  • High-performance edge computing, coupled with advanced sensors, to give scientists software-defined sensors  leveraging new ML algorithms at the edge.
  • A cloud-based development environment for students and scientists to write edge-to-cloud software pipelines
  • A programmable infrastructure to test the limits and capabilities of machine learning and deep learning for future intelligent sensor designs.
  • Replicable technology, policy, and process for federated growth beyond initial partner instruments.


Recent News