1. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/1Ad Hoc and Sensor Networks – Roger Wattenhofer – Positioning Chapter 8 TexPoint fonts used in EMF. Read the TexPoint manual before you delete this box.: AAAA A A A A

2. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/2Ad Hoc and Sensor Networks – Roger Wattenhofer – Lecturer: Rainer Mautz Geodetic Metrology and Engineering Geodesy Institute of Geodesy & Photogrammetry, ETH Zurich HIL D43.2, Hönggerberg, Wolfgang-Pauli-Str. 15, CH-8093 Zurich +41 44 633 24 85 http://www.geometh.ethz.ch/people/rmautz

3. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/3Ad Hoc and Sensor Networks – Roger Wattenhofer – Stratovolcano Sakurajima, island 77 km 2 1117 m height extremely active, densely populated monitored with levelling, EDM, GPS Landsat image, created by NASA Application: Natural Hazards Monitoring (e.g. Volcanos) Volcanoes experience pre-eruption surface deformation Coverage: cm – dm over 10 km 2 Spatially distributed monitoring for early warning system WLAN positioning system with densely deployed location aware nodes GPS WLAN

4. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/4 Rating Area maturity Practical importance Theoretical importance First steps Text book No apps Mission critical Not really Must have Sensor data without position information are normally worthless

5. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/5Ad Hoc and Sensor Networks – Roger Wattenhofer – Overview Sensor Types Positioning Systems Methods for Positioning

6. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/6Ad Hoc and Sensor Networks – Roger Wattenhofer – High Level Functional Elements Positioning and Tracking Standards Pro- cessing Legal Issues Communi- cation Ranging Positioning Requirements Parameters Accuracy AvailabilityUpdate Rate Measurement Rate Integrity Coverage Inter- faces Safety and Security User Requirements for Positioning Functionality

7. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/7Ad Hoc and Sensor Networks – Roger Wattenhofer – availability:100% of the time timeliness:realtime reliability:no failures hybrid systems: to be avoided local installations:none accuracy:mm - cm coverage:global all environments: indoors: household, office & factory outdoors: urban & rural dynamic User Requirements:

8. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/8Ad Hoc and Sensor Networks – Roger Wattenhofer –  Principle (trilateration, triangulation, signal strength)  Signal (Radio Frequencies, Light Waves, Ultrasound, RFID, Terahertz)  Environment (indoor, outdoor, urban, rural, remote)  Active / passive sensors  Accuracy (μm – km)  Application (industry, surveying, navigation) Classification of Positioning Systems:

9. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/9Ad Hoc and Sensor Networks – Roger Wattenhofer – Positioning Systems

10. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/10Ad Hoc and Sensor Networks – Roger Wattenhofer – Sensors used in Geodetic Networks 1.optical sensors (motorised total station, digital level) 2.GNSS sensors (that allow phase measurements) 3.geotechnical sensors (relative measurements using extensometer, inclinometer, plumbing, DMS, hydrostatic levelling) 4.meteo sensors (thermometer, pressure sensor, air moisture sensor) key aspects & issues: -physical dimensions of the sensor (size, weight) -additional devices for data collection -format of generated data -measurement rate -connection to computer network -electrical power consumption and voltage -price

11. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/11Ad Hoc and Sensor Networks – Roger Wattenhofer – Positioning Sensors 1. optical sensors – motorised total stations not yet specialised for that purpose (still can be used manually) problems: sensitive electronic, optical sight (fog), price, need for prisms S6/8

12. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/12Ad Hoc and Sensor Networks – Roger Wattenhofer – Totalstation Network automatic total station: a 1-node monitoring network from: Heunecke, Geosensornetze im Umfeld der Ingenieurvermessung

13. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/13Ad Hoc and Sensor Networks – Roger Wattenhofer – Positioning Sensors 1. optical sensors – motorised total stations Mt.Terri Felslabor - Messeinsatz – Stephan Schütz

14. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/14Ad Hoc and Sensor Networks – Roger Wattenhofer – Sensors used in Engineering Geodesy Networks 1. optical sensors – digital levels -restricted to 40 m and horizontal views -restricted to monitor a single staff

15. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/15Ad Hoc and Sensor Networks – Roger Wattenhofer – Sensors used in Engineering Geodesy Networks 2. GNSS sensors – handheld & geodetic receivers Why do geodetic receivers reach mm-accuracy?

16. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/16Ad Hoc and Sensor Networks – Roger Wattenhofer – Sensors used in Engineering Geodesy Networks 1.Use of both frequencies L1& L2 2.Phase measurements 3.DGPS (relative measurements) elimination of ionospheric delays from: WIDE University (School of Internet) http://www.soi.wide.ad.jp/class/20050026/slides/01

17. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/17Ad Hoc and Sensor Networks – Roger Wattenhofer – Sensors used in Engineering Geodesy Networks 2. GNSS sensors – OEM-Boards (Original Equipment Manufacturer) -receivers better suited for monitoring: low power 6 - 30V, 0.2 watts -like a PC card, without antenna, data and power interfaces -cheap “GPS chips” not sufficient for precise DGPS & phase measurements  requirement for high computing power

18. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/18Ad Hoc and Sensor Networks – Roger Wattenhofer – Material[dB]Factor [-] Glass1 - 40.8 – 0.4 Painted Glass100.1 Wood2 - 90.6 – 0.1 Roofing Tiles / Bricks5 - 310.3 – 0.001 Concrete12 - 430.06 – 0.00005 Ferro-Concrete29 - 330.001 – 0.0005 GNSS Attenuation of building materials (L1 = 1500 MHz) Stone (1997) Signal Strength in Decibel Watt of GNSS Satellites Environment[dBW] Satellite+14signal strength delivered from satellite Outdoors-155unaided fixes OK for standard receivers Indoors-176decode limit for high sensitive receivers Underground-191decode limit for aided, ultra-high sensitive receivers

19. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/19Ad Hoc and Sensor Networks – Roger Wattenhofer – Sensors used in Engineering Geodesy Networks 3. Geotechnical Sensors -only one-dimensional, mostly local, relative measurements -high effort for installation (cables, adaption to the object, e.g. plumbing shafts, inclinometer pipes) -analogue signals  A/D converter, logging unit

20. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/20Ad Hoc and Sensor Networks – Roger Wattenhofer – Sensors used in Engineering Geodesy Networks 4. Meteorological Sensors important for correction of other sensors Meteo-Station WS-2308: T, p, wind direction, wind speed, participation, time reference using DCF-77 Mainflingen

21. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/21Ad Hoc and Sensor Networks – Roger Wattenhofer – Sensors used in Engineering Geodesy Networks 5. Digital Cameras IP-camera has its own IP-address. It can be used as a webcam that does not require a direct connection to a computer Cameras will play a more important role! IP-Camera NC 1000-W10: -detection of movements, -usable in the night -WLAN capability -detection of sounds

22. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/22Ad Hoc and Sensor Networks – Roger Wattenhofer – Positioning (Localization) Task: Given distance or angle measurements or mere connectivity information, find the locations of the sensors Anchor-based –Some nodes know their locations, either by a GPS or as pre-specified. Anchor-free –Relative location only. Sometimes called virtual coordinates. –Theoretically cleaner model (less parameters, such as anchor density) Range-based –Use range information (distance or angle estimation). Range-free –No distance estimation, use connectivity information such as hop count. –It was shown that bad measurements don’t help a lot anyway.

23. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/23Ad Hoc and Sensor Networks – Roger Wattenhofer – Measurement Methods for Positioning Angle estimation Angle of Arrival (AoA) –Determining the direction of propagation of a radio-frequency wave incident on an antenna array. Distance estimation Received Signal Strength Indicator (RSSI) –The further away, the weaker the received signal. Time of Arrival (ToA) or Time Difference of Arrival (TDoA) –Signal propagation time translates to distance. –RF, acoustic, infrared and ultrasound.

24. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/24Ad Hoc and Sensor Networks – Roger Wattenhofer – Lateration Methods Multilateration: Initial anchor Step 1 : Step 2 : Step 3 : becomes anchor Iterative Multilateration: Collaborative (n-hop) Multilateration: Positioning Based on Range Measurements

25. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/25Ad Hoc and Sensor Networks – Roger Wattenhofer – Ambiguity problem when creating the smallest rigid structure Positioning Based on Range Measurements

26. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/26Ad Hoc and Sensor Networks – Roger Wattenhofer – Solving flip ambiguity in the presence of noise Positioning Based on Range Measurements

27. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/27Ad Hoc and Sensor Networks – Roger Wattenhofer – Expansion of a rigid structure by multilateration Positioning Based on Range Measurements

28. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/28Ad Hoc and Sensor Networks – Roger Wattenhofer – iGPS transmitter and sensor during a test in a tunnel AoA Measurement: iGPS

29. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/29Ad Hoc and Sensor Networks – Roger Wattenhofer – PrincipleOutdoorIndoor Real- time AccuracyRange Signal Frequency Data Rate MarketCost TOA angular measurements 0.1 – 0.2 mm 2 - 50 mRF40 Hzin progresshigh Key design:  two or more fixed transmitters  rotating fan-shaped laser beams  infrared signal  various sensors detect arrival times  position determination with spatial forward intersection graphic from Metris AoA Measurements: iGPS “laser resection”

30. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/30Ad Hoc and Sensor Networks – Roger Wattenhofer – Locata: Terrestrial pseudolite transceivers Alternative Positioning Systems Picture from Barnes et al. (2003) 6thIinternational Symposium on Satellite Navigation Technology, Melbourne, Australia

31. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/31Ad Hoc and Sensor Networks – Roger Wattenhofer – SystemPrinciple Outdoo r Indoor Real- time AccuracyRange Signal Frequency Data Rate MarketCost Locata TOA, lateration 2 mm static 1 cm RTK, 2 - 3 kmRF1 Hz in progress high Locata – Key Parameters: (+) RTK: 1 – 2 cm deviations at 2.4 m/s (+) signal magnitude stronger than GNSS (+) indoors dm Problem: multipath (low elevation) Picture from J. Barnes, C. Rizos, M. Kanli, A. Pahwa „A Positioning Technology for Classically Difficult GNSS Environments from Locata“, IEEE Conference, San Diego California, 26 April 2006

32. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/32Ad Hoc and Sensor Networks – Roger Wattenhofer – TDoA: Ultrasound Systems – Crickets, Active Bat, Dolphin SystemPrinciple Outdoo r Indoor Real- time AccuracyRange Signal Frequency Data Rate MarketCost Cricket TOA, lateration  1 – 2 cm10 multrasound1 Hz develop ment low Active Bat TOA, lateration  1 – 5 cm1000 m 2 ultrasound75 Hznomoderate DOLPHIN TOA, lateration  2 cm room scale ultrasound20 Hznomoderate Cambridge University

33. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/33Ad Hoc and Sensor Networks – Roger Wattenhofer – SystemPrinciple Outdo or Indoo r Real- time AccuracyRange Signal Frequenc y Data Rate MarketCost Cricket TOA, lateration  1 – 2 cm10 m ultrasoun d 1 Hz develop ment low Active Bat TOA, lateration  1 – 5 cm1000 m 2 ultrasoun d 75 Hznomoderate DOLPHI N TOA, lateration  2 cm room scale ultrasoun d 20 Hznomoderate Problems:  dependency on temperature  maximal range  deployment of reference beacons  multipath  reliability  interference with other sound sources TDoA: Ultrasound Systems – Crickets, Active Bat, Dolphin

34. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/34Ad Hoc and Sensor Networks – Roger Wattenhofer – Positioning based on Signal Strength All signals can be used: WLAN, Ultrasound, RF, GPRS, etc. Problems:  reliability  accuracy SystemPrinciple Outdoo r Indoor Real- time AccuracyRange Signal Frequency Data Rate MarketCost SonitorRSSI, Cell ID  m-level15 multrasound0.3 Hzyeslow RFID Signal Strength  dm-m20 m RF, 866 MHz nolow USC Robotics Research Lab RSSI in sensor networks: not for geodetic (reliable, precise) applications!

35. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/35Ad Hoc and Sensor Networks – Roger Wattenhofer – Auto-Positioning Algorithm Mobile listener collects distance measurements redundancy = R – 3(B + P) + 6, R = observed ranges, B = fixed beacons P = listener positions Example below: redundancy = 0 (R = 24, B = 4, P = 6) Ceiling Listener (moving) Beacons (static)

36. Ad Hoc and Sensor Networks – Roger Wattenhofer –8/36Ad Hoc and Sensor Networks – Roger Wattenhofer – Auto-Positioning Algorithm Procedure: a)collection of range measurements b)grouping of listener positions c)detection of gross errors d)distance matrix e)completion of approximate distance matrix by MDS or lateration f)computation of approximate positions in local system g)refinement by global and local optimisation B1B1 B2B2 B3B3 B4B4 P1P1 P2P2 P3P3 P4P4 P5P5 P6P6 B1B1 B2B2 not observedobserved B3B3 B4B4 P1P1 P2P2 P3P3 not observed P4P4 P5P5 P6P6