Ubiquitous Wireless Sensors Kris Pister Founder Chief Technologist

Ubiquitous Wireless Sensors Kris Pister Founder & Chief Technologist Dust Networks kpister@dustnetworks. com (510) 813 -3878 Network Services Configurable Data Filter/Control Analog I/O Digital I/O Serial Port

Dust Networks v Reliable, low-power wireless networks for sensing Select Dust OEM Customers Select OEM End Users 2

Wireless Sensor Networks 1. Vision 2. Applications 3. Technology & Standards 4. Future 3

The Vision § Access to MORE DATA § You can’t control what you don’t measure § Measure and improve: - Efficiency, yield, lifetime, up-time, safety, … 4

The Requirements § Deploy devices ANYWHERE § The data must be RELIABLE § No battery changes for many YEARS 5

Extreme Wireless Screen captures from “Extreme Wireless Applications” by Emerson Process Management on You. Tube http: //www. youtube. com/watch? v=JFz. RNz. Hq. Fz. Y These are all Wireless. HART devices built by Emerson and networked by Dust 6

7

8

9

10

11

12

13

14

15

Emerson offerings, 2007 (Dave Farr presentation) 16

Why use wireless sensors? § Reduced cost of installation § Condition-based maintenance § Increased production § Reduced operation costs 17

Chevron’s Richmond Refinery 1 km 18

Richmond Refinery Wireless Umbrella Next • Fence monitoring • H 2 S, VOC • Location 3 km 2, 90% coverage 19 19

Condition-based maintenance Failure probability § With no sensing, need to repair early § With sensing, repair near average time 20

Wheeling-Pittsburgh Steel Need to monitor temp, coolant, lubrication Hot slag defeated wired solutions 5% improvement in productivity (reduced maintenance) 21 21

Smart Rail: Amsted Rail v Multiple sensors per car: bearing temp, hatches, … v Installer toolset: v “A sledgehammer and a blowtorch” 22

Bearing Failure High Cost 23

Reduced processing times Completion probability § With no sensing, need to cook longer § With sensing, mean process time = mean required time 24

Lime Kiln at Pulp & Paper Mill • Rotating lime kiln • Need to monitor temperature • 5% throughput improvement (reduced process time) 25 25

Data Center: Reduced operating costs v Vigilent provides data center energy management service by deploying hundreds of sensors and control points in the data center v No wires, no interruption to data center operations v E. g. Verizon is saving > 55 M k. Wh annually in 24 major data centers, reducing greenhouse gases by 66 M pounds/year 26

Building Monitoring: IBM and The Met v Hundreds of sensors monitor temperature and humidity to help preserve medieval art v Historic building: requires no wires, no construction v Builds a 3 D "climate map" for predictive analytics & trending 27

Smart City: Parking, Streetline Networks Wireless sensor node 28

Finding Parking 29

Parker app 30

Hydrology: Berkeley § Shaver Lake, CA (1900 m) - 60 WSN nodes - 300+ Sensors 31

Sensor Networking Standards § TSMP – Wireless HART (IEC 62591), ISA 100. 11 a, 802. 15. 4 e § Zigbee § IPv 6 – 6 Lo. WPAN 32

Time Synchronized Channel Hopping Primer Network & Security Management IEEE 802. 15. 4 Radio v v Time slotted and synchronized for low power and scalability Full mesh and channel hopping for wire-like reliability (>99. 99% ) Every node can run on batteries (or energy harvester) for 10+ years Incorporated into IEC 62591 and IEEE 802. 15. 4 E standards 33

50 motes, 7 hops 3 floors, 150, 000 sf >100, 000 packets/day 34

Time Synchronization § Timestamps available to application - RMS <0. 1 ms with Wireless HART - RMS <0. 01 ms with Smart. Mesh IP 35

Zig. Bee Primer v IEEE 802. 15. 4 radio v Line-powered routers, low power edge nodes v Single frequency v Zig. Bee ‘Green Power’ is for batteryless sensors and switches for building & residential energy management 36

Smart Grid 15. 4 G Concentrating solar SE 1. x, SE 2. x, … 37

6 Lo. WPAN Primer Network & Security Management IEEE 802. 15. 4 Radio v IPv 6 end-to-end through the internet v Same reliable low power mesh from Dust 38

802. 15. 4 Markets and Standards Zig. Bee/Green. Power 6 Lo. WPAN Zig. Bee SE 2. 0 (in meter) Zig. Bee RF 4 CE INFRASTRUCTURE IEC 62591 Cities Rail Water Zig. Bee 6 Low. PAN Various 39

802. 15. 4 radio current vs. sample date RX Current 0 d. Bm TX Current CEL MSP 430 +CC 2420 Ember TI Freescale Ember TI Jennic Dust Networks 40 40

Smart. Mesh SPOT Asset Management System & Location Engine Network Manager Locn: Room 327, west wall Fixed Battery Powered Mote 27. 2 m 22. 5 m 40. 1 m Mobile Mote 17. 8 m 23. 2 m Sensor v All the advantages of ‘no-wires’ installation v Good accuracy (<4 m, 90% CDF) and reliability 41

Energy scavenging: no batteries required! Zigbee sensor IPv 6 routers En. Ocean: mechanical Nextreme: thermal Solar Micropelt: 42 thermal

Scavenging Wireless HART Products ü Battery ü Vibration ü Battery ü 4 -20 m. A loop ü Solar ü Battery ü 4 -20 m. A loop ü Thermal 43

Emerson and Dust § 2004 -- partnership with Process Management - Created Wireless. HART, IEC 62591 § Dozens of wireless sensors to date - Rosemount, Fischer, Liebert, … - Voltage, current, time, temperature, flow, vibration, … § 2010 RFP from Emerson Network Power - 10 responses - Dust Networks selected as strategic wireless sensor networking technology provider across Network Power product groups in April 2011 § IPv 6, RTLS, … 44

Networked by Dust Over 10, 000 networks in 120 countries oil refineries, museums, mines, chemical plants, city streets, sierra snow, trains, steel mills, pharma plants and data centers 45 © 2011 Dust Networks, Inc.