IJkdijk Clay and mud applications of advanced fibre and wireless networks Robert Meijer
IJkdijkClay and mud applications of advanced fibre and wireless networks
Robert Meijer
I will not tell about
Robert Meijer …
• Professor in ICT / sensor networks University Amsterdam
• Sr strategist @ TNO
• Nuclear physicist• Network research
www.gsi.de
Wedstrijd 300 400 500m
6
SOFTWARE MANIFESTATION of networks
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
PSCRS
X
PIN/PDC
PPBAC
Telco
infrastructure
OC
namespace WebApplication10
{
/// <summary>
/// Summary description for WebForm10.
/// </summary>
public class WebForm10 : System.Web.UI.Page
{
…
private void ButtonCall_Click(object sender, System.EventArgs e)
{
LambdaSwitch LS1 = new LambdaSwitch(o.114.1,o.113.2,
o.114.2,o.114.1);
FastStore FS1 = new FastStore(100,”Mbyte”);
Switch S1 = new Webtel();
Telco T1 = new Telco(“KPN”);
Router R1 = new router(T1, 139.63.63.1,139.63.63.2,
139.63.63.3);
….
}
}
}
MATH networksnetworks manifested in Mathematica
6/21/2009- 7 -
Needs["WebServices`"]
<<DiscreteMath`Combinatorica`
<<DiscreteMath`GraphPlot`
Print["The following methods are available from the
NetworkComponent:",InstallService["http://
localhost:3000/network_service/service.wsdl"]];
The following methods are available from the NetworkComponent:
{GetAllLinks,GetAllElements,NetworkTokenTransaction}
Initialisation:
n = GetAllElements[];
e = GetAllLinks[];
nids = Apply[Union,e];
Print["Network elements: ", n];
Print["Number of ports found: ", Length[nids]];
Network elements: {bigvirdot,virdot}
Number of ports found: 16
NE Discovery:
Topology and visualization of the
shortest path:
nodePath = ConvertIndicesToNodes[
ShortestPath[ g,
Node2Index[nids,"192.168.3.4"],
Node2Index[nids,"139.63.77.49"]],
nids];
Print["Path: ", nodePath];
If[NetworkTokenTransaction[nodePath, "green"]==True,
Print["Committed"], Print["Transaction failed"]];
Path:
{192.168.3.4,192.168.3.1,139.63.77.30,139.63.77.49}
Committed
Shortest path, transactions, use of
tokens:
8
Adaptive networks
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
Calient
0.11A.1
0.11A.2
0.11A.3
0.11A.4
0.13A.3
PIN/PDC
PPBAC
0.13A.2
R1.forward(139.63.65.0,0.0.0.0);
…
path p = new path();
networkElelement ne = new networkElement;
pathCollection Paths =
new pathCollection(UniversalLambdaAddress1,
UniversalLambdaAddress2);
{
for each p in paths
{
#BEGIN MANUAL TRANSACTION T
try {
for each ne in p {
T.register(ne);
ne.reserve(“25-mrt-2004 1030h”, “1000s”);}
#T.commit();}
catch {t.rollback());
…
PSC
Telco
infrastructure
OC
NETWORK PROGRAMS tell
the network AS A WHOLE how to be a service1 CREATE Route
2 FROM
3 SELECT query(
4 SELECT * FROM
NetworkElement ne1
WHERE ne1.identifier = F
5 )
6 TO
7 SELECT query(
8 SELECT * FROM
NetworkElement ne2
WHERE ne2.neighbors.size = 1
9 )
10 USING "DijkstraShortestPath"
1 <?xml version="1.0" encoding="ISO-8859-1"?>
2 <xmi:XMI xmi:version="2.0" xmlns:xmi="http://www.omg.org/XMI"
3 xmlns:ac="ac">
4 <ac:ApplicationComponent identifier="forwarder" revision="1">5
6 <handlers handlerType="MSG_RECEIVED"
7 functionName="handleMessage" priority="5" />8
9 <functions name="main">
10 <statements name="LABEL">
11 <arguments type="String" value="loop" />
12 </statements>
13 <statements name="SLEEP" />
14 <statements name="JUMP">
15 <arguments type="String" value="loop" />
16 </statements>
17 </functions>18
19 <functions name="handleMessage">
20 <statements name="POP">\
21 <arguments type="String" value="packetDestination" />
22 </statements>
23 <statements name="POP">
24 <arguments type="String" value="packetToken" />
25 </statements>
26 <statements name="POP">
27 <arguments type="String" value="packetType" />
28 </statements>29
30 <statements name="PUSH">
31 <arguments type="String" value="packetToken" />
32 </statements>
33 <statements name="PUSH">
34 <arguments type="String" value="1209804020" />
35 </statements>
36 <statements name="EQ" />37
38 <statements name="IF_TRUE_GOTO">
39 <arguments type="String" value="match" />
40 </statements>
41 <statements name="RETURN" />42
43 <statements name="LABEL">
44 <arguments type="String" value="match" />
45 </statements>
46 <statements name="PUSH">
47 <arguments type="String" value="packetType" />
48 </statements>
49 <statements name="PUSH">
50 <arguments type="String" value="packetToken" />
51 </statements>
52 <statements name="PUSH">
53 <arguments type="String" value="C" />
54 </statements>
55 <statements name="PUSH">
56 <arguments type="String" value="upvn" />
57 </statements>
58 <statements name="SEND" />
59 <statements name="RETURN" />60
61 </functions>62
63 </ac:ApplicationComponent>
64 </xmi:XMI>
Application Domain
NETWORK COMPILER is compiled by ITSELF
A
ANC
NC
C
D
6
4
11
2
4
3
1
5
2
8
6
10
9
7
11
a
bc
d
e
f
Programming the ant heap.
Interactive Networks
Periodic Networks with SunSpots
I will tell about
Applications of PROGRAMABLE NETWORKS
Programmable antennae
Software controlled radio
Field Programmable Gate Array• Multiple “calculated” CPU• “Calculated OS”• VMWARE• Dedicated applications
How to program anOPTIMAL Application specificNetwork?
But not aboutInterplanetary Telecommunications
But about Clay
And about Mud
And how ICT prof becomes dike expert
Robert Meijer
TNO & University of Amsterdam
IJkdijkThe development of an early warning system for dike failures
Not about the Ike Dike
Field lab
DIKES?
What makes, from an ICT perspective, smart dykes so interesting?What is the IJkdijk fieldlab?
Because one is allowed to destroy something? Ja!
23
Nice pictures? Of course.
CostsNew dikes: 500-5000€/mDike upgrade: 500-5000€/m
Dikes:Netherlands: 15 000 kmWorld: 1 000 000 km
Value protected:Rotterdam: 750 000 000 €
40 km
3000-2M kilometers @ 100 bytes/s/m => 2,4 Gbit/s - 1,6 Tbit/s
Rotterdam, only the largest dikes are indicated
Over dimensioning (e.g. more clay!) does not help, is technically impossible or cannot be paid
10000KM OF SENSORS, COMPUTERS, SUPERCOMPUTERS AND TELECOMONLINE
EVERYTHING IS SOFTWARE EVERYTHING IS ONLINE
AI MONITORS, CONFIGURES, REPAIRS
Application of advanced networks:
Too weak, not too low
Basics of a smart and active dike– Sensing components
• sensitive everywhere along, below the dike
• Acoustics
• Fiber-Bragg optical fibers
– Acting components• Drainage
• Pumps
– Energy supply• Windmills
– Sensor telecommunication system
– Command & control• Multiple authorities
• Multiple command and control centers
– Finance
Command & Control
Pump
Sea
Information SpaceInformation Space
Information Space
SSS
Joint ObservatoriesJoint Observatories
Joint Observatories
Early Warning System Components
• Information Space
– shared acces to data
• Command & Control
– Facilitate human interaction
• Observatories
– Federated command & control centers
– E.g. Rhine Observatory
Sink
Sensors &Actuators
Command & Control
Telecommunication
TNO, Deltares, UvA
50 Companies
80 dikes will be destroyed
Robert Meijer, Rudolf Strijkers, Leon Gomans,
Cees de Laat32
Experimental
area
Experimentation has to be learned: test dike jan 2007
8
7
6
54
3
2
1
0
-1
-2
-3-4
-5
-6
-7
-8
-9
-11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
8
7
6
54
3
2
1
0
-1
-2
-3-4
-5
-6
-7
-8
-9
-11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
..
..bassin
Water in/outletsFibre Braggmovement
Fibre Braggmovement
Invertedpendulum
Seismic (acoustic)
Macro stability dikeChannel dike
Channel
Macro stability experiment23-27 september 2008
Dike Survey
• 20+ FBG cables
Hansje Brinker BV – laser scan
Reflection intensity map
Δdistance (top) and calculated Δdistance sensitivity of radar satellite (ERS, Envisat) (bottom)
Satellite radar interferrometry: mm, however measures displacements under more unfavorable circumstances: 22deg angle, once in 6 days
Laser: 1.5cm
Laser
Satellite
Alert Solutions - GeoBeads
Temperature -40 +100 deg, pressure 0-2500 cm H20, humidity inclination -90+90 degrees, 3D acceleration 1/20000g
Ready to go, but when?
Macro stability experiment:Several groups in the world calculated well enough the
time of collapse
• An early warning system – can be created
– practical aspects of system become important
3000-2M kilometers @ 100 bytes/s/m => 2,4 Gbit/s - 1,6 Tbit/s
Sensor systems summary
• All systems detect huge differences from normal just before the time of collapse
• All systems detect smaller differences before collapse
– Without simulations and calculations (in progress) no proof of early warning capacities
– Without more experiments no phenomenological proofs
• Systems detect humans, machines, helicopters, animals
NOW: GET THE ICT DEVELOPMENTSSTARTED
For 10000km dikes
No world model -> learn it. AI
Difference
Sensordata
Historical data
Analysis DecisionEnsemble of likely states
WarningSensor data
Decision Warning
Detailed survey dike Models Historical information
Ijkdijk starts with
Organize this as an online service
MonitorS
SS
Control CentreSSS
Control Centre
Scenario Computation VisualisationVisualizationMonitor
Information SpaceInformation Space
Decision SupportDecision Support Scenario Computation
Public Information
Internet
Dike sensor network
Landscape model
Remote monitoring
Water height monitor
Decision support
Public Information
Legacy earlywarning system
Grid/cloud resourcesGoogle, Microsoft, Yahoo, Amazon, 3Tera, IBM, …
Urban Flood Early Warning System Services
OnlineInformation
Super Computers
Programmable Networks