SYS 0: CENS Systems Research Overview
Jeff Burke, Deborah L. Estrin, Ramesh Govindan, Richard Guy, Mark H Hansen, William Kaiser, Eddie Kohler, Todd D. Millstein, Jens Palsberg, Miodrag Potkonjak, Mani B. Srivastava, John D. Villasenor · eScholarship (California Digital Library) · 2006
Center for Embedded Networked Sensing Systems Research Overview Jeff Burke, Deborah Estrin, Ramesh Govindan, Richard Guy, Mark Hansen, William Kaiser, Eddie Kohler, Todd Millstein, Jens Palsberg, Miodrag Potkonjak, Mani Srivastava, John Villasenor Introduction Deployments to Instruments Addressing challenges of field deployed systems Initial challenges: adaptive self configuration in communication sensing Difficult environment, challenging connectivity issues, heterogeneity FROM… Spatial redundancy Planned long-term deployments Self-configuring autonomy Post-deployment analysis Pre-calibration Trusted communications One user TO ALSO INCLUDE… Actuated placement Rapid short-lived deployments Interactive exploration Live analysis feedback Integrity monitoring Safety and robustness Multiple users Future challenges: Multi-modal, multi-scale observing instruments Architecture of observing systems, supporting rapid deployment and reconfiguration, and supporting multiple and mobile users Selected Research in Posters Distributed Acoustic Arrays • Integrates CENS research in signal processing, localization, time synchronization, and hardware and software platforms • Nodes self-localize (X, Y, Z, Θ) in difficult 3D terrains with foliage, background noise, adverse weather • High accuracy acoustic source localization Node 108 Node 104 Distributed Image Sensing • Vision is a rich sensing modality – Context, shape, motion, presence, color… • Novel Cyclops platform for large-scale networks of low- power low-resolution cameras – Benefits of diverse perspectives and proximity • Energy optimized algorithms, coordinated actuation, dynamic reconfiguration • Deployment of vision network at James Reserve Programming and System Architecture Architecting for Manageability and Robustness Programming High-data Rate Applications Amplifier and Microphone Bio- High-Integrity Design Sensor Networks in the Presence of Failures software faults resource ownership node failures, packet loss calibration, drift, outliers or Interface Platform & System Integrity (debugger, static checkers, run-time fault isolation) Measurement & Data Integrity (detection, remediation, recalibration) 12-bit ADC TelosB mote Control Data USB Interface to PC Network Integrity (network fault isolation) i686 µServer Macroprogramming Distributed Applications Node Crashed Yes Node Crash No Node Rebooted Yes No No Node Reboot Have Neighbors Yes Have Route • Tests on all passively and actively collected statistics (Ovals) • Root causes assigned based on test results (Rectangles) Yes Rx Pkts No Neighbors No No Route No Bad Node Inbound Path No Bad Node Outbound Path Yes Tx Pkts Yes Bad Sink Inbound Path UCLA – UCR – Caltech – USC – CSU – JPL – UC Merced