permasense real-time permafrost monitoring · • tree based routing towards data sink – no...
TRANSCRIPT
PermaSense Real-time Permafrost MonitoringJan Beutel, ETH Zurich
PermaSense
• Interdisciplinary geo-science and engineering collaboration• Consortium of several projects, start in 2006• Fundamental as well as applied research
– Long-term, high-quality sensing in harsh environments– Better quality data, obtained online– Measurements that have previously been impossible– Enabling new science, answering fundamental questions related to
decision making, natural hazard early-warning
• More than 35 people, 17 PhD students
[J. Beutel et al: X-Sense: Sensing in Extreme Environments. Proc. Design, Automation and Test in Europe (DATE 2011).V. Wirz et al: Temporal characteristics of different cryosphere-related slope movements in high mountains. Proc. Second World Landslide Forum, 2011.B. Buchli, F. Sutton and J. Beutel: GPS-equipped Wireless Sensor Network Node for High-accuracy Positioning Applications. Proc. EWSN 2012.]
Matter Valley: Detecting Large-Scale Mass Movements
Our patient does not fit into a
laboratory
So the laboratory has to go on the
mountain
Our Field Sites: Precision Scientific Instruments
Saastal
Miniature Low-Power Wireless Sensors
• Static, low-rate sensing (120 sec)• Simple scalar values: voltages, resistivity, digital sensors• 4-5 years operation (~150 μA avg. power)• ~0.1 Mbyte/node/day• 8+ years experience, ~1’276’035’519 data points
7
Ruggedized for Alpine Extremes
[A. Hasler et al: Wireless Sensor Networks in Permafrost Research -Concept, Requirements, Implementation and Challenges. Proc. 9th International Conference on Permafrost (NICOP), 2008.]
A base station collects and relays
the data to the valley
[B. Buchli, F. Sutton, J. Beutel and L. Thiele: Dynamic Power Management for Long-Term Energy Neutral Operation of Solar Energy Harvesting Systems. Proc. SenSys 2014.B. Buchli, F. Sutton, J. Beutel and L. Thiele: Towards Enabling Uninterrupted Long-Term Operation of Solar Energy Harvesting Embedded Systems. Proc. EWSN 2014M. Keller, J. Beutel and L. Thiele: The Problem Bit. Proc. DCOSS 2013 ★ Best Paper Award ★B. Buchli, D. Aschwanden and J. Beutel: Battery State-of-Charge Approximation for Energy Harvesting Embedded Systems. Proc. EWSN 2013.M. Keller, J. Beutel and L. Thiele: How Was Your Journey? Uncovering Routing Dynamics in Deployed Sensor Networks with Multi-hop Network Tomography. Proc. SenSys 2012.]
PermaSense Core System Architecture
[J. Beutel et al: PermaDAQ: A Scientific Instrument for Precision Sensing and Data Recovery under Extreme Conditions. Proc. IPSN/SPOTS 2009.]
WLAN Long-haul Communication
• WLAN (802.11a) backbone using directional links• Leased fiber/DSL from Zermatt Bergbahnen AG to mountaintop
Online Data Management & Access• Global Sensor Network (GSN)
– Data streaming framework from EPFL (K. Aberer)– Organized in “virtual sensors”, i.e. data types/semantics– Hierarchies and concatenation of virtual sensors enable on-line processing– Dual architecture translates data from machine representation to SI values,
adds metadata
Data from field site is received by the private GSN server “as is” and storedin a primary database.
Data is passed on to a public GSN server where it is mapped to metadata (positions, sensor types, calibration) and converted to convenient data formats.
Data is available for download and analysis using external tools.
firewall
NETWORKED EMBEDDED SYSTEMS RESEARCH
Sensor Node Hardware • Shockfish TinyNode184
– MSP430, 16-bit, 8MHz, 8k SRAM, 92k Flash– LP Radio: SX1211 @ 868 MHz
• Waterproof housing and connectors
• Sensor interface board– Interfaces, power control– Temp/humidity monitor– 1 GB Flash memory
• 3-year life-time– Single Li-SOCl2 battery, 13 Ah– ~300 µA power budget
[J. Beutel et al: PermaDAQ: A Scientific Instrument for Precision Sensing and Data Recovery under Extreme Conditions. Proc. IPSN/SPOTS 2009.]
Challenge: The Physical Environment
• Strong daily variation of temperature– −30°C to +40°C– ΔT ≦ 20°C/hour
• Impact on – timing, energy availability, fatigue, SOFTWARE, …
Impact of Environmental Extremes
[J. Beutel et al: X-Sense: Sensing in Extreme Environments. Proc. Design, Automation and Test in Europe (DATE 2011).]
• Tighter guard times increase energy efficiency• Software testing in a climate chamber
– Clock drift compensation yields ± 5ppm
• Validation of correct function
Ultra Low-Power Data Gathering• Dozer ultra low-power data gathering system
– Beacon based, 1-hop synchronized TDMA– Optimized for ultra-low duty cycles– 0.167% duty-cycle, 0.032mA (@ 30sec beacons)
• Dynamic adaptation– Back off randomization for diversity– Jitter adaptation over multiple hops– Account for long-term loss of connectivity
17
[Burri, N., von Rickenbach, P., & Wattenhofer, R.: Dozer: Ultra-Low Power Data Gathering in Sensor Networks. Proc. IPSN 2007.]
contention window
Dozer System (i)
• Tree based routing towards data sink– No energy wastage due to multiple paths
• TDMA based link scheduling– Each node has two independent schedules – No global time synchronization
• The parent initiates each TDMA round with a beacon– Enables integration of disconnected nodes– Children tune in to their parent’s schedule
time
beacon
beacon
activation frame
18
Dozer System (ii)
• Parent decides on its children data upload times– Each interval is divided into upload slots of equal length– Upon connecting each child gets its own slot– Data transmissions are always ack’ed
• No traditional MAC layer– Transmissions happen at exactly predetermined point in time – Collisions are explicitly accepted– Random jitter resolves schedule collisions
time
jitter
slot 1 slot 2 slot n
data transfer
19
Global vs. Local Time Sync
In cases where no network-wide time synchronization is available Global time sync not available for network protocol control Implications on data usage
Solution: Elapsed time on arrival Sensor nodes measure/accumulate packet sojourn time Base station annotates packets with UTC timestamps Generation time is calculated as difference
•2011/04/14 10:03:31 – 7 sec•= 2011/04/14 10:03:24
4 sec
1 sec
2 sec
4 sec6 sec 7 sec
a
b
c
20
[M. Keller, J. Beutel and L. Thiele: Multi-hop network tomography: path reconstruction and per-hop arrival time estimation from partial information. ACM SIGMETRICS Performance Evaluation Review Volume 40, Issue 1, p. 421-422, 2012.M. Keller, J. Beutel and L. Thiele: How Was Your Journey? Uncovering Routing Dynamics in Deployed Sensor Networks with Multi-hop Network Tomography. Proc. SenSys 2012.
FlockLab Testbed @ ETHZ• 4 Target Architectures
− Tmote Sky− TinyNode− Iris−Opal
• 30 Node Testbed− Ethernet/WLAN backbone− In- & Outdoor
21[R. Lim et al: FlockLab: A Testbed for Distributed, Synchronized Tracing and Profiling of Wireless Embedded Systems. Proc. IPSN 2013.]
Synchronized Power and State Traces from Distributed Nodes
PSender
PReceiver
Sender Radio TX
Sender CPU
Test Case: Synchronized Glossy Floods
[F. Ferrari et al:: Efficient Network Flooding and Time Synchronization with Glossy. Proc. IPSN 2011. ★ Best Paper Award ★F. Ferrari, Marco Zimmerling, Luca Mottola and Lothar Thiele: Low-Power Wireless Bus. Proc. SenSys 2012. ★ Best Paper Award ★]
Net
wor
k di
amet
er [h
ops]
Average synchronization error
PulseSync
FTSP
Glossy
1 μsGPS receiver
FTSP with high resolution timer0
10
20
30
2 μs
TPSN
Ethernet (PTP)
Head-to-Head Comparison of Network Time Synchronization
24
Short / longTATS
TATS3 x
4 x – 7 x
Dynamic
[Roman Lim, Balz Maag and Lothar Thiele: Time-of-Flight Aware Time Synchronization for Wireless Embedded Systems. Proc. EWSN 2016.]
200 ns22 hops
Comparison Using GPS Reference Time
25
Reference node (1)
Nodes equipped with GPS (7)
Comparison Using Different Network Topologies
26
Parameters
• 1 s synchronization interval• Regression over 80 samples• Duration: 1 h
Short
Long
Dynamic
182 m / 22 hops
283 m / 22 hops
Head-to-head comparison
27
0 0.5 1 1.5 20
20
40
60
80
100TATSshort/long/dynamic
Error to reference [µs]
Frac
tion
[%]
Glossydynamic PulseSync long
PulseSync short
PulseSync dynamic
WIRELESS L1-DGPS SENSORS
L1 GPS Sensor Developments
GPS Logger• Large-scale, early
access data
GPS CoreStation• Experimentation, variable use
Wireless GPS Sensor• Fully integrated, low-power
29
GPS Sensor Cluster Example: Dirruhorn
30
Base Station
• Network clusterhead– Aggregates all sensor
network data– Moves data to long-haul
network (WLAN, GPRS)
• GPS position reference• Hardware
– Embedded Linux– WLAN/GPRS– 12V solar power– Other sensors
31
Wireless GPS• Wireless communication
– Sensor network cluster– 868 MHz low-power– Up to 5km range
• Sensors– ublox LEA-6T GPS– 2-axis inclinometer– Ambient temp/hum/battery
• Data logger function– Local 2GB data buffer
• Remote configurable – Duty-cycle– Sampling rate
32
Capturing Millimeter Scale
Process Dynamics
[Wirz, V. et al: Estimating velocity from noisy GPS data for investigating the temporal variability of slope movements, Nat. Hazards Earth Syst. Sci., 14, 2503-2520, 2014.V. Wirz et al: Short-term velocity variations at three rock glaciers and their relationship with meteorological conditions. Earth Surface Dynamics, 4, 1, p. 103-123, 2016.]
Real-time Experimentation at Valley-Scale
34
Access to Real-time Data for Early Warning Decision-making
Bielzug Debris Flow, June 2013• Critical natural hazard event• Herbriggen partial village evacuation• Closure of road and railway to
Zermatt
Längschnee, Fall 2014• Constructive measures securing rock
boulders above Herbriggen• Extension of sensor coverage in
collaboration with authorities
35
Technology TransferPERMOS Continuous GPS Pilot
• Pilot program to make L1-DGPS sensors available on PERMOS partner field sites
• First sensor installation in summer 2013, extensions in 2014, 2015(Valais: Herbriggen Bielzug, Breithorn + Längschnee, Grächen Distelhorn + Ritigraben, Saas-Balen Gruben + Jäggihorn, Wysse Schije, Randa Grossgufer)
Hanging glacier break-off, Weissmies, Sept. 2014
PERMOS site Largario rock glacier, Sept. 2014 Avalanche protection, Randa, Nov. 2014
36
Monitoring of Weissmies Hanging Glacier
• Multi-modal Monitoring of Hanging Glaciers – Ground-based radar (Geopraevent)– High-resolution imaging– In-situ wireless GPS
installed from long-line
-10
-5
0
5
10
15
20
25
30
35
40
26.05.2015 15.07.2015 03.09.2015 23.10.2015 12.12.2015 31.01.2016 21.03.2016
Triftgletscher 3D Velocity [cm/d]
Position 2 v_3d
Position 3 v_3d
10 per. Mov. Avg. (Position 2 v_3d)
10 per. Mov. Avg. (Position 3 v_3d)
37[L. Preiswerk et al.: Monitoring unstable parts in the ice-covered Weissmies northwest face. Proc. 13th Congress INTERPRAEVENT 2016.]
• ETH Zurich– Computer Engineering and Networks Lab– Geodesy and Geodynamics Lab– Micro and Nanosystems
• University of Zurich– Department of Geography
• EPFL– Distributed Information Systems Laboratory
• University of Basel– Department Computer Science
Interested in more?
http://www.permasense.ch