Milestones, Feedback, Action Items Power Aware Distributed Systems Kickoff August 23, 2000.

Milestones, Feedback, Action Items

Power Aware Distributed Systems Kickoff

August 23, 2000

Power Aware Distributed Systems

Impact Power-aware algorithms, sensor node RTOS,

and middleware will reduce sensor network aggregate energy requirements >1000X.

This capability will extend sensor network power dynamic range to span from prolonged (months) quiescent operation to “get me the information now at any cost”.

Power instrumentation of existing low-power sensor node provides baseline by which PAC/C tools and technology will be measured.

Goals Algorithms. Develop power-aware algorithms for

cooperative signal processing that exploit sensor data locality, multi-resolution processing, sensor fusion, and accumulated intelligence.

Protocols. Design a distributed sensor network control middleware for power-aware (P-A) task distribution and hardware/software resource utilization migration.

Compilers/OS. Create sensor node RTOS to manage key resources – processor, radio, sensors.

Systems. Identify hardware power control knobs and readable parameters and make them available to the sensor node power-aware RTOS.

Milestones [FY/Q] P-A RTOS scheduling on research platform [01/Q1]. Instrumentation board for research platform [01/Q1]. Compressed image transmission (Laplacian Pyramid) [01/Q1]. SensorSim simulation tool with P-A extensions [01/Q4]. Tool for power-aware RTOS kernel synthesis [02/Q4]. Deployable platform with P-A control “knobs” [02/Q4]. P-A network resource allocation DP field demo [03/Q2]. RP w/ sensor-triggered activation & low power sleep [03/Q3]. High-res multi-look image classification demo [03/Q4].

Extending dynamic power range for distributed sensor

networks.

Sensor Node Hardware Control Knobs and Power

Aware RTOS

Cooperative Signal Processing

Sensor Network Middleware

Power Aware Distributed Systems

Q: How can you extend the dynamic power range of sensor networks from quiescent months of monitoring to frenetic minutes of activity?

Architectural Approaches Power Aware Research Platform Testbed Deployable Power Aware Sensor Platform

Middleware, Tools, and Techniques Power Aware Resource Scheduling in RTOS Techniques for Network-Wide Power Management

Power Aware Algorithms Multi-Resolution Distributed Algorithms

Architectural Approaches

Instrument a state-of-the-art sensor node to understand power consumption in current systems. Where can we expect significant power tradeoff? Which knobs have the greatest dynamic range? What baseline will we use for comparison?

Develop power reconfigurable communications module. Adapts parameters such as error control, equalization,

data rate, and noise figure in real time according to channel state. Leverage existing FPGA-based Rockwell radio.

FY02/Q1

FY01/Q1

Power Aware Sensor Node RTOS

Identify hardware knobs for radio/processor modules that can be altered dynamically by a power aware RTOS. Find knobs in existing commercial deployable sensor nodes. Introduce new knobs into research platform for eventual tech

transfer into deployed commercial systems.

Identify readable parameters (power, BER, signal strength, battery, etc.) that can be provided to power-aware RTOS. Identify available parameters in deployed sensor nodes and

instrument parameters in future PAC/C research platform modules.

Provide operating system extensions for power management, task scheduling, and task control on individual sensor nodes. Extend open API to expose and take advantage of available power

aware capabilities on multiple research and deployed platforms.

FY02/Q4

FY02/Q1

FY01/Q1

FY01/Q1

Power Aware Resource Scheduling

Traditional RTOS schedulers manage processor from perspective of time. Key issue for power aware sensor node RTOS is to manage all resources – processor, radio, sensors.

Use both static scheduling and dynamic scheduling. Fixed priority preemptive scheduling (e.g. rate monotonic). Dynamic priority preemptive scheduling (e.g. earliest deadline-first).

The basic approach will exploit slack time in the schedule to shutdown a resource or to operate it at a lower-power lower-speed setting.

A crucial observation that we offer is that missed deadlines in many computation tasks on a sensor node are no different than noise in the radio or sensor: it would result in a radio or sensor packet being dropped.

FY02/Q4

FY03/Q2

Techniques for Network-Wide Power Management

Design a distributed sensor network control middleware for power-aware task distribution and hardware/software resource utilization migration. Extends existing low-power routing protocols developed in SensIT.

Incorporate power trade-off analysis tools into the SensIT platform emulator for power aware application development and scenario simulation for sensor networks. SensorSim simulator allows analysis with a larger system than is

practical with real nodes. Use SensIT topographical map GUI to visualize network power

consumption behavior and analyze power aware techniques against replayed scenarios from SensIT field experiment data sets.

FY02/Q3

FY01/Q4

Multi-Resolution Distributed Algorithms

Develop power-aware algorithms for cooperative signal processing that exploit sensor data locality, multi-resolution processing, sensor fusion, and accumulated intelligence. Adapt Laplacian Pyramid techniques for the efficient transmission

and processing of images at multiple resolutions. Multi-resolution image target classifier based on neural networks. Multi-resolution, hierarchical sensor cueing including acoustic and/or

low-res imaging sensor cueing simulator demo. Directed high-res multi-look image classification/validation demo.

FY01/Q1

FY02/Q1

FY02/Q4

FY03/Q4

Financial Profile

FY00 FY01 FY02 FY03 Total

$620,791 $1,481,410 $432,154 $450,000 $2,984,355

Rockwell $150,000 $ 407,000 $100,000 $136,000 $ 793,000UCLA $150,000 $ 450,000 $129,000 $ 729,000ISI $320,791 $ 624,410 $203,154 $314,000 $1,462,255

Feedback

Action Items