An Ultra Low Power System Architecture for Sensor

  • Slides: 23
Download presentation
An Ultra Low Power System Architecture for Sensor Network Applications Mark Hempstead, Nikhil Tripathi,

An Ultra Low Power System Architecture for Sensor Network Applications Mark Hempstead, Nikhil Tripathi, Patrick Mauro, Prof. Gu-Yeon Wei, Prof. David Brooks Division of Engineering and Applied Sciences Harvard University Cambridge, MA

Overview • Wireless sensor networks (WSN) are constrained by energy consumption • Goal: Average

Overview • Wireless sensor networks (WSN) are constrained by energy consumption • Goal: Average power consumption <100 µW enables energy scavenging methods • Our architectural approach: • • Holistic approach Event-driven architecture Modular hardware accelerators Fine-grain power management • In the implementation phase 2

Outline • What are sensor networks? • Project motivation and design constraints • Event-driven

Outline • What are sensor networks? • Project motivation and design constraints • Event-driven architecture • Performance and power estimates • Conclusion and future work 3

Sample Application Space • Monitoring Apps • Structural/Earthquake/Weather/Habitat monitoring • Building/Border/Battlefield detection • Road/traffic

Sample Application Space • Monitoring Apps • Structural/Earthquake/Weather/Habitat monitoring • Building/Border/Battlefield detection • Road/traffic monitoring • Medical Apps • Long-term health monitoring • Untethered Pulseox Sensors • Business Applications • Supply Chain Management • Expired/Damaged Goods Tracking • Automatic Checkout Systems 4

Example App: Great Duck Island • Great Duck Island (GDI), Maine - (UC Berkeley)

Example App: Great Duck Island • Great Duck Island (GDI), Maine - (UC Berkeley) • • • Gather temp, humidity, IR readings from Leach's Storm Petrel burrows and weather station motes Determine occupancy of nests to understand migration patterns Total of 150 nodes deployed in 2003, over 650, 000 observations taken • Performance Requirements are Low • Samples taken and transmitted once every 5 min • Power consumption limited lifetime of deployment Single Hop Network Multi-Hop Network R. Szewczyk et al. An Analysis of a Large Scale Habitat Monitoring Application. ACM Conference on Embedded Networked Sensor Systems (Sen. Sys), 2004. 5

Example Sensor Network Node Low Power Low Throughput Tiny. OS for Event Driven Wireless

Example Sensor Network Node Low Power Low Throughput Tiny. OS for Event Driven Wireless Communication and Adhoc Networking Programmable CPU Battery Operated Interface to Various Sensors Small Form Factor Mica 2 Mote – Designed by UC Berkeley, Manufactured by Crossbow 6

Energy is the primary limitation CPU Mode Current @3 V Radio Mode Current @3

Energy is the primary limitation CPU Mode Current @3 V Radio Mode Current @3 V Active 8. 0 m. A Receive 7. 0 m. A Idle 3. 2 m. A Transmit Min Power 3. 7 m. A Standby 216 µA Transmit Max Power 21. 5 m. A Power-save 110 µA Sensor Board 0. 7 m. A • Mica 2 Power Consumption Measured by component • Not the complete picture, how is power consumed in an application? V. Shnayder, M. Hempstead, B. Chen, G. Werner-Allen, M. Welsh. Simulating the Power Consumption of Large-Scale Sensor Network Applications. (Sen. Sys'04). 7

Application-level Power Analysis Total energy consumption per component of “Surge”, a multi-hop routing application,

Application-level Power Analysis Total energy consumption per component of “Surge”, a multi-hop routing application, run for 60 sec on the Mica 2 mote. Due to General Purpose architecture of CPU Requires software overhead to run Tiny. OS Can be decreased at application and protocol levels however this requires more CPU computation Design Goal: Average Power consumption of < 100 µW to enable energy scavenging methods. Where should design energy be focused to decrease energy consumption? 8

Regular Application Behavior Sense and Transmit Abstract View Timer Interrupt Collect Sensor Data Prepare

Regular Application Behavior Sense and Transmit Abstract View Timer Interrupt Collect Sensor Data Prepare Message Send Radio Message Receive and Forward Example - GDI Every 5 min Burrow Occupancy - infrared - humidity - Pack data in packet - Calculate checksum - wait for acknowledgement Message Arrives Decode Message Search Routing Table Resend Radio Message 9

Flexibility/Programmability Key goals of our architecture • • • General Purpose CPU Remove Software

Flexibility/Programmability Key goals of our architecture • • • General Purpose CPU Remove Software Overhead OUR SYSTEM Retain Programmability ASIC Energy Efficiency Event-driven computation Hardware accelerators for power-efficiency Exploit regular operations Optimize for sensor net workloads Modular design Fine-grain power management 10

Abstract view of architecture Shared Memory Radio Transceiver General Purpose Event Microcontroller Processor Slave

Abstract view of architecture Shared Memory Radio Transceiver General Purpose Event Microcontroller Processor Slave Blocks Sensors 11

Detailed view of architecture Radio Addr/Data Event Processor Interrupt Power Ctrl SRAM Sensors Addr/Data

Detailed view of architecture Radio Addr/Data Event Processor Interrupt Power Ctrl SRAM Sensors Addr/Data System Bus Micro Controller Message Processor Data Filter Timer • Regular events mapped solely to EP and slaves • Micro Controller included for irregular events • Slaves provide application specific HW • All resource usage is explicit 12

Event Processor • Interrupts invoke EP interrupt service routines • 8 instructions • 4

Event Processor • Interrupts invoke EP interrupt service routines • 8 instructions • 4 power control/control transfer • 4 read/write/transfer data to devices 13

App. Example: Sense + Transmit System Initialization/Reprogram Collect Sensor Data Prepare Message Micro Controller

App. Example: Sense + Transmit System Initialization/Reprogram Collect Sensor Data Prepare Message Micro Controller Sensors Radio Addr/Data Event Processor Interrupt Power Ctrl SRAM System Bus Timer Interrupt Message Processor Data Filter Timer Send Radio Message Configuration written to memory and timer 14

Example: Sense + Transmit (2) Collect Sensor Data Prepare Message Send Radio Message Micro

Example: Sense + Transmit (2) Collect Sensor Data Prepare Message Send Radio Message Micro Controller Radio Addr/Data Event Processor Sensors Addr/Data Interrupt Power Ctrl SRAM System Bus Timer Interrupt Message Processor Data Filter Timer Pseudo Code <timer intaddr>: SWITCHON <sensor> SWITCHON <message proc> TRANSFER <reading size> <sensor addr><message proc addr> SWITCHOFF <sensor> WRITEI <ctrl_wrd> <message proc> TERMINATE; 15

Example: Sense + Transmit (3) Collect Sensor Data Prepare Message Send Radio Message Micro

Example: Sense + Transmit (3) Collect Sensor Data Prepare Message Send Radio Message Micro Controller Radio Addr/Data Event Processor Interrupt Power Ctrl SRAM Sensors Addr/Data System Bus Timer Interrupt Message Processor Data Filter Timer Pseudo Code <message proc mesg. ready intaddr>: SWITCHON <radio> TRANSFER <mesg size> <message proc> <radio> SWITCHOFF <message proc> WRITEI <ctrl_wrd> <radio> TERMINATE; 16

Example: Sense + Transmit (4) Collect Sensor Data Prepare Message Micro Controller Radio Addr/Data

Example: Sense + Transmit (4) Collect Sensor Data Prepare Message Micro Controller Radio Addr/Data Event Processor Send Radio Message Interrupt Power Ctrl SRAM Sensors Addr/Data System Bus Timer Interrupt Message Processor Data Filter Timer Pseudo Code <radio, message sent intaddr> SWITCHOFF <radio> TERMINATE; 17

Example: Sense + Transmit (5) Collect Sensor Data Prepare Message Send Radio Message Micro

Example: Sense + Transmit (5) Collect Sensor Data Prepare Message Send Radio Message Micro Controller Radio Addr/Data Event Processor Interrupt Power Ctrl SRAM Sensors Addr/Data System Bus Timer Interrupt Message Processor Data Filter Timer System Idle 18

Implementation • Process technology study (see paper) • Does Moore’s Law help us? •

Implementation • Process technology study (see paper) • Does Moore’s Law help us? • Leakage power increasing concern • Tradeoff active power and leakage power • Architectural enables low power circuit techniques • Fine-grain power management – VDD gating • Simple Circuit Implementation • • Synchronous design VDD roughly 2 VT Performance Target: 100 k. Hz Possible to use less common circuit design styles (subthreshold, asynchronous) 19

Initial Results • Developed performance model for system architecture in System. C (~8 K

Initial Results • Developed performance model for system architecture in System. C (~8 K lines of code) • GP microcontroller, event processor, slave blocks, radio • Power Model • VHDL for Event Processor + Key Blocks • Custom design (SRAM, CAM) • 0. 25 µm Process Technology • Workload Analysis and early comparison to other architectures included in the paper 20

Performance Comparison Roughly 10 x cycle-reduction justifies 100 KHz clock speed 21

Performance Comparison Roughly 10 x cycle-reduction justifies 100 KHz clock speed 21

Power estimates Unknown blocks: GP microcontroller, busses, off-chip interfaces 22

Power estimates Unknown blocks: GP microcontroller, busses, off-chip interfaces 22

Conclusion/Future work • Wireless Sensor Networks provide unique opportunities for low power, low throughput

Conclusion/Future work • Wireless Sensor Networks provide unique opportunities for low power, low throughput design • Architecture meets design goals • Less than 100 µW average power consumption • Event Processor provides event handling in HW • HW slaves provide application specific processing for regular tasks • Fits sensor network application characteristics • Implementation phase of first chip • Stay Tuned! 23