challenge: ultra-low-power energy- harvesting active networked tags (enhants) authors maria...

34
Challenge: Ultra-Low-Power Energy-Harvesting Active Networked Tags (EnHANTs) Authors Maria Gorlatova, Peter Kinget, Ioannis Kymissis, Dan Rubenstein, Xiaodong Wang, Gil Zussman Presenter Velin Dimitrov

Upload: bryce-caldwell

Post on 23-Dec-2015

216 views

Category:

Documents


1 download

TRANSCRIPT

Challenge: Ultra-Low-Power Energy-Harvesting Active

Networked Tags (EnHANTs)

Authors Maria Gorlatova, Peter Kinget, Ioannis Kymissis, Dan

Rubenstein, Xiaodong Wang, Gil Zussman

Presenter Velin Dimitrov

• Small• Flexible• Self-Reliant (energy)• Attached to non-networked objects

Books, Clothing, Produce, etc.

EnHANTS

• Between sensor networks and RFID tags in capabilities

EnHANTs

• RFID -> Identify• EnHANTs -> Actively search

• Enables continous monitoring and querying

• Misplaced book detector• Detect people in disasters

Why EnHANTs?

• “enablers for the Internet of Things”

• Today computers—and, therefore, the Internet—are almost wholly dependent on human beings for information. Nearly all of the roughly 50 petabytes (a petabyte is 1,024 terabytes) of data available on the Internet were first captured and created by human beings—by typing, pressing a record button, taking a digital picture or scanning a bar code. Conventional diagrams of the Internet ... leave out the most numerous and important routers of all - people. The problem is, people have limited time, attention and accuracy—all of which means they are not very good at capturing data about things in the real world. And that's a big deal. We're physical, and so is our environment ... You can't eat bits, burn them to stay warm or put them in your gas tank. Ideas and information are important, but things matter much more. Yet today's information technology is so dependent on data originated by people that our computers know more about ideas than things. If we had computers that knew everything there was to know about things—using data they gathered without any help from us—we would be able to track and count everything, and greatly reduce waste, loss and cost. We would know when things needed replacing, repairing or recalling, and whether they were fresh or past their best. The Internet of Things has the potential to change the world, just as the Internet did. Maybe even more so.

Internet of Things

• Network – Multihop• Ultra Low Power – nJ/bit• Harvest Energy• Energy Adaptive• Exchange messages (IDs)• Trasmit 1-10m• Thin and flexible

Capabilities

• Scheduling of query responses of RFID

• Sensor networks• IEEE 802.15.4a

Based on Impulse Radio UWB Ranging 26 Mb/s Backward Compatible

Existing Research

• Dynamic activation of energy-harvesting sensors

• Outsourcing packet retransmissions of energy restricted nodes

• Texas Instruments solar energy harvesting

Existing Research

• Temperature Differences• Electromagnetic Energy• Airflow• Vibrations• Solar• Motion

Energy Harvesting

• Solar – 100 to 0.1 mW/cm^2• Office/retail/lab settings brighter

than residential• Single and polycrystalline solar

cells 10-20% efficient• Efficiency depends on energy

availability

Rigid Solar Cells

• Organic semiconductors• Constant efficiency w.r.t brightness• 1.5-1% efficiency

Flexible Solar Cells

• 10 cm^2 organic semiconductor cell

• Outdoors -> 10 mW• Single bit -> 1nJ• Achievable data rate is 10 Mb/s• Indoors -> 10 uW• Achievable data rate is 10 Kb/s

Solar Energy Use Example

• Energy captured mechanically• Users have some control• PVDF – 4 uJ/cm^2 per 1.5%

deflection

Piezoelectric Harvesting

• 10 cm^2 of material• Strained 60 times/sec• Generates 2.4 mW• At 1nJ/bit -> 2.4 Mb/s

Piezoelectric Example

• Compact/efficient, low self discharge

• Ability to measure and control• Rechargeable batteries

Thin-film batteries -> flexible Supplied voltage must exceed internal

chemical potential to charge Voltage upconversion

Energy Storage

• Capacitors• Receive any voltage exceeding • Cycle time• More difficult to add charge as the

capacitor reaches capacity• Self discharging• Battery -> 1000 J/cm^3• Capacitor -> 1-10 J/cm^3

Energy Storage

• C -> Energy storage capacity• E -> Current energy level• r -> Energy charge rate• e -> Energy consumption rate

Energy Abstraction

• UWB impulse radio (IR)• Current state of art (2009)

50 pJ/bit transmit 500 pJ/bit receive 100 Kb/s to 1 Mb/s

Communciations

• Conventional modulated systems require transmitter active for entire duration of signal transmission

• UWB uses pulses so transmit requires very little power

• No difference in listening to media and receiving information

Communications Paradigm Shift

• Accurate clocks consume too much energy

• Ultra low power ring oscillators can be used but drift significantly

• Protocol redesign?• Clock redesign?

Challenge: Inaccurate Clocks

• Spend more energy when it is available

• More accurate clocks Help other tags synchronize

• Run radios more to discover nodes• Open research area

High Power Mode

• Tags do not maintain contact• Energy use is decided on a per-tag

basis• Transmit/Receive duty cycling• Transmit only mode for low energy

Independent State - Communications

• If tags receive and transmit cycles align, they can pair

• Similar to low power modes of IEEE 802.11 and Bluetooth

• Keep-alives are the bursts• Burst frequency <-> clock drift

Paired State - Communications

• Tags exchange (C,E,r,e)• Low energy dictates how much of

this info can be exchanged• How this occurs exactly is key to

minimizing energy

Communicating State

• Leverage the environment• Synchronization via a channel

always open• Can be used to influence joint

energy decisions• Higher layer challenges

Polling/Pushing Security/Privacy

Multiple EnHANTs

• Deterministic model• Periodic energy source

Office lights

• Highest energy consumption rate:

Energy Management and Optimization

Economic Production Quantity

• 10 cm^2 solar cell with 1% efficiency

• Dim shelf -> 50 mW/cm^2• Maximum energy consumption 2.08

uW• Translates to 2.08 Kb/s throughout

day at 1nJ/bit

Example

• Order point, order quantity model

Stochastic Model

• More random, fewer known parameters Storage capacity

• Cannot manufacture energy, can manufacture goods

• Both supply and demand are stochastic

• Dependency of energy means tags are like a network of factories

Differences with Inventory

• Custom UWB IR transceiver on MICA2

• Synchronized OOK• 2.18 nJ/bit• 18 kbps at 1-3 meters• Bit error rate under 0.001• CSMA• Energy harvesting module ->

battery, solar cell, and charging circuit

EnHANT Prototype

EnHANT Prototype

Testbed

•  ISO/IEC 14543-3-10• 30 m range• Wireless light switches• Solar• 14 byte packets• 125 kbps• $1.4 B sales in 2013

EnOcean