15829B18849B95811A19729A InternetScale Sensor Systems: Design and Policy

1
15-829B/18-849B/95-811A/19-729AInternet-Scale
Sensor Systems Design and Policy

• Lecture 3 Code Overview Project Ideas

2
Outline

• Code Overview
• Project Ideas

3
Code Availability

• http//www.intel-iris.net
• Code accessible via CVS
• What is CVS ? version control system
• Useful for group access to source code
• Multiple people can edit the same file but
  conflicts may occur
• Anonymous CVS ? cant check in code

4
Mailing-lists

• Class members
• irisnet-course_at_intel-iris.net
• Iriscode developers
• irisdev_at_intel-iris.net
• Managed by majordomo_at_intel-iris.net

5
Using Laptop

• Laptop setup
• irisnet and root account ? passwords irisnet
• Feel free to customize account but dont change
  password
• IrisNet source code checked out in
  irisnet/IRISNET
• Redhat 8.0
• Boot with camera attached
• Say configure at all the configuration prompts
• Log in as irisnet
• Start X server
• Network configuration
• System settings ? network
• Add ? wireless ? lucent ? SSID CMU
• Activate
• Start browser with outside CMU URL ? register
  wireless

6
Building IrisNet

• cd IRISNET
• export CVS_RSHssh
• cvs login
• password anonymous
• cvs up -d -P
• ./install
• Should build without error
• Try test suite
• cd /IRISNET/TestSuite ./testall.sh

7
Components

• SA
• C-based
• Base SA code
• OpenCV
• CamStream webcam viewer
• OA
• Java-based
• Base OA code
• Apache Xindice XML database
• Logger
• NAM network simulation animator
• Applications

8
SA Architecture
Sensor Feed
Senselet
Senselet
Shared Memory
Network
Senselet
SA Daemon
9
SA Sensor Feed

• Read from sensor and place info in shared memory
• One process per sensor
• Sensor readings are stored in circular buffer
• Provides access to historical measurements as
  space permits

10
SA Starting the Sensor Feed

• cd SA/IrisWebcam
• ./runwebcam.sh to use real camera
• ./loadImage to load one image
• ./loadImage images/pX.jpg 75
• X number of full spots 0..6
• 75 shared memory key for IrisNet
• ./loadImages to cycle through files in images
  subdir
• ./showImage shows the current image

11
SA CamStream

• CamStream provides an interface to webcams and a
  bunch of useful tools
• E.g., webcam capture/record
• Needs gcc 2.95 (not installed on laptops) to
  compile completely
• Gcc 3.2 only compiles the few libraries critical
  to IrisNet

12
SA Senselet

• Senselet
• One senselet per sensor service
• Each senselet runs as a separate process
• Senselet typically does
• 1) take sensor data from shared memory
• 2) processes data to produce result
• 3) Transmits result to configured OA
• 4) Results may be stored back in shared memory

13
SA Running Senselet

• ./SAControl 127.0.0.1 3456 S 127.0.0.1 6789
  parking executes parking detection senselet
• Need to first run Util/Portsink/portsink 6789
  to discard data
• Can load new images to see changes in result
• Can use similar technique to test your code for
  mini-project

14
SA Example Senselet

• cd Applications/parking/SA
• parking.cc contains parking space senselet
• Main function called Start(FilterParameter param)
• Typical structure of main function
• Open shared memory
• cvCreateImageHeader and cvSetImageData calls
  retrieve video image
• openCV calls to process image data
• Send data to the OA

15
SA SA Daemon

• Manages the execution of senselets
• Accepts commands from OAs to start/stop execution
• Text-based command protocol (port 3456)
• L name
• uploads the code for a senselet and places it in
  filesystem
• S
• starts execution of a senselet and transmits
  output to specified OA
• U
• stop execution of a senselet

16
SA SA Daemon

• Starting
• cd SA/IrisSA/src
• ./runsa
• Controlling
• Normally controlled by OAs
• cd Util/SAControl
• ./SAControl to manually send commands

17
OA - Architecture
Xindice
OA Daemon
Network
18
OA Xindice

• What does it do?
• Local XML database engine
• Stores XML documents, handles Xpath 1.0 (mostly
  select and update)
• Interfaces
• Command line ? enter Xpath
• Java API
• What doesnt it do well
• Xquery, XSLT
• XSLT processing ? done in separate Java library
• Image storage
• How do we use it
• Each service has a single XML document that
  stores sensor readings and static metadata

19
OA Starting Xindice

• cd packages/xml-xindice-1.0
• ./xindice.server start
• Make sure that config/xindice.pid does not exist
• export XINDICE_HOMEpwd
• cd packages/xml-xindice-1.0/bin
• Run ./xinidiceadmin to control database

20
OA OA Daemon

• Listens on port 6789
• Handles processing/forwarding of queries
• Uses XSLT to query Xindice database
• Manages partitioning of database across nodes
• DNS entries
• Reaction to load

21
OA OA Daemon Messages

• Query query
• Process the XPath query, id identifies source
  query made at root, fragment identifies this
  particular subquery
• Reply
• Answer to a query
• Take ownership
• Start a new OA
• Delegate ownership query
• Split an OA
• Delete database
• Load data from SA
• Used by SAs to add to database
• Update DNS
• Use dynamic DNS

22
OA OA Daemon

• Running
• cd OA/IrisOA
• Edit oa.cfg if needed
• make run
• Controlling
• make runcl
• Type help to see interface
• Running a Service
• cd Applications/parking/OA
• Look at demodataone.xml
• Follow handout instructions to load xml document

23
Logging

• What are the OAs/SAs doing?
• Log all messages ? currently uses a central
  server
• Challenge how to order and playback log messages
• Lamport clocks to ensure causal order
• ns-2 network animator (nam) used to playback log
• What is nam?

24
Logging

• Running the logger
• cd logger/IrisLogger/src
• runnam.sh
• Running NAM
• cd logger/nam-1.0a11a_iris
• ./nam ../IrisLogger/src/logfile.6
• Currently logging must be carefully tuned to the
  application ? manual steps involved
• Name and layout of nodes in animation
• Conversion of logfile node names to nam nodes
  identifiers

25
Outline

• Code Overview
• Project Ideas

26
Projects

• 2 or 3-person groups
• IrisNet-based
• Mini-project provides intro to IrisNet
• Expectations
• 10 page report presentation
• Workshop quality results
• Project ideas
• Will be posted on Web pages

27
Projects (cont.)

• Generous donation from Intel
• 20 laptops cameras
• 10 desktops
• Each project group will receive 2 laptops and 3
  cameras
• Handed out TODAY in CMCL lab 3604 Wean Hall
• Only to enrolled students, other students will
  need to wait
• No replacements take good care of them
• No grade until you return them in good condition!
• Interesting deadlines
• Sensys Apr 8th
• Mobicom workshops June middle

28
Applications Monitoring Interesting Places

• At lunch places
• At grad lounge foosball table
• NSH Atrium
• Find me a free conference rooms
• Tennis/raquetball/etc. court avail
• Desk area/terminal room availability
• Car theft monitor
• Call/email owner if thief is near car!
• Traffic monitoring
• Auto traffic (traffic jams)
• Throughput of area (cars, people) - speed trap

29
Applications Object Tracking

• Person tracking
• Where did my child/pet go
• Advisor/advisee avoidance
• Location of my/someone else's car
• Inventory control

30
Other Service Types

• Audio sensor-based applications
• Identify language spoken
• Identify individual based on voice
• Triggered sensor applications
• Record picture when car horn is heard
• Network monitoring as an IrisNet application
• Intrusion detection system

31
Other Service Types Environment Modeling

• Smart room support
• 3d positioning based on multiple sensors
• Virtual reality tour
• Take the sensor input to create parts of a
  virtual reality
• Weather monitoring

32
Applications General Challenges

• Image processing
• How do you identify objects of interest?
• Person/object counting techniques
• Motion detection based
• Identifying the same object in different views
• Handoff of tracking data from one sensor to next

33
Applications General Challenges

• Privacy issues
• Access control to OA database
• How does a person opt out of a surveillance
• Query processing
• What type of data schema for sensor readings?
• How are historic queries handled
• Data mining
• Long term trends e.g., in parking space
  availability or waiting lines

34
Infrastructure

• Incorporating new sensor types in IrisNet
  infrastructure
• Sensor motes how do you modify the SA
  programming model to accommodate motes?
• Others mobile sensors, public webcams, smart
  camera phones
• Making IrisNet infrastructure robust to failures
• Replicating part of database
• How to handle dynamic vs. static parts of
  database
• Handling failures in routing of queries e.g.,
  due to caching?
• Distributed monitoring of system and on-demand
  collection of logs

35
Infrastructure Energy Constraints

• Power management for local processing, e.g., how
  accurately can we predict energy consumption?
• What is the energy tradeoff between local
  processing of data vs. sending the data to do
  remote processing?
• Running a CPU, a sensor device, or a
  communication device in sleep mode can save
  energy considerably. An approach for saving power
  is thus to operate our devices mostly in sleep
  mode, and have them turn on periodically.
• How much accuracy does our sensor data
  loosegiven some duty cycle?
• Can we find more intelligent scheduling
  algorithms than simply uniform duty cycles (on
  during 100ms each second)?
• What about using low-energy sensors to schedule
  morepower-hungry sensors?

36
Infrastructure

• Location-based services
• How could we incorporate location-specific
  information in our sensors?
• How could we establish or verify the location of
  a sensor?
• Using video landmarks
• Image Processing
• Can we anonymize raw sensor feed e.g., by
  blurring people in image

37
Infrastructure Policy/Security

• How can a user ensure that a video feed has not
  been misused?
• E.g., monitor the output XML, perform test
  queries on system
• Assuming an open environment (i.e., no central
  certificate authority everybody trusts), how can
  we establish trust between unrelated entities?
• How do we ensure that sensor data we receive is
  trustworthy and originates from the correct
  sensor?
• If anyone can make queries/program sensors, how
  do we protect the system from a denial-of-service
  attack?

38
Infrastructure

• DNS
• Performance evaluation
• DHTs as an alternative
• Code safety and resource allocation on SAs/OAs
• How are services protected from each other
• How are nodes protected from malicious services
  that consume resources
• Storage use on SAs
• Support for historic sensor reading retrieval
• Impact on result sharing

39
Infrastructure

• Push vs. Pull of data
• Impact on monitoring of failed nodes
• Impact on caching/response time for queries
• Handling aggregate fields in OA database

40
Next Lecture

• Image processing
• OpenCV library
• Assigned reading
• TBA