ESNA Technical Platform

1
ESNA Technical Platform
European Sensor Network Architecture (ITEA
project 05023)
Project Review Stockholm, Sweden Nov 14, 2007

• Adam Dunkels, PhD (SICS)
• WP1 leader
• Sensor network infrastructure OS and networking

2
Executive Summary

• The problem
• Sensor network software development exceedingly
  difficult
• Our solution
• A down-to-earth expressible abstraction framework
• Expected results
• Higher quality of ESNA sensor networks
• Shorter time to market

3
Technical challenges
4
Sensor network programming today
Network/transport/application/
Deluge
VRR
Flush
CTP
Trickle
Application 1
AODV
RBP
GRE
S4
ETX/le
Application 2
ZigBee
Z-MAC
Crankshaft
X-MAC
MAC
B-MAC
S-MAC
SCP
6lowpan
802.15.4
Link
TR1001
nRFxx
CC2420
CC1100
5
What we want to achieve
Application 2
Application 1
Best-effort multi-hop unicast
Reliable bulk download
Best-effort network dissemination
Reliable bulk dissemination
Reliable flooding
Reliable data collection
802.15.4
X-MAC
6lowpan IPv6
Wireless HART
6
Challenging properties

• Cost requirements
• Distributed
• Embedded
• Wireless
• Potentially large scale
• Changeable environment, system
• Resource-challenged devices
• Energy (batteries)
• Memory (512 b 10kb RAM, 48kb ROM)
• Processing power (2 MHz)

7
What we want to achieve (again)
Application 2
Application 1
Best-effort multi-hop unicast
Reliable bulk download
Best-effort network dissemination
Reliable bulk dissemination
Reliable flooding
Reliable data collection
802.15.4
X-MAC
6lowpan IPv6
Wireless HART
8
Challenges

• Programming abstractions
• Many low-level mechanisms
• Resource consumption
• Performance
• Making use of link layer mechanisms
• Full-system optimization

9
ESNA an Expressible Sensor Network Abstraction
framework

• Communication abstractions rather than protocols
• Choose protocol based on performance requirements
• In-network data storage abstractions
• Data placement based on performance requirements
• Node-level reprogrammable data processing

10
ESNA principles

• Separation of concerns
• Layered modularity
• Separation of protocol mechanisms and protocol
  headers
• Separation of system mechanisms and system
  configuration

11
The ESNA project
12
ESNA is timely

• The field of sensor networks is ready
• Sensor network architecture a hot topic
• ESNA has a unique position
• OS competence (Contiki)
• Application competence
• Several applications
• System developer competence

13
On-going related projects

• SNAP (Berkeley)
• http//snap.cs.berkeley.edu/
• Tenet (USC)
• http//enl.usc.edu/projects/tenet/
• TinyCubus (Stuttgart)
• http//www.ipvs.uni-stuttgart.de/abteilungen/vs/a
  bteilung/forschung/projekte/tinycubus/start

14
Implications of our work

• Critical part of every sensor node out there
• 100 M devices per year!
• Open source software, guidelines
• Potential for huge impact
• ESNA/Contiki is C, the de-facto standard embedded
  language
• Important research to be made

15
The role of WP1

• Technical platform used by other WPs
• Enabling software
• Basic building blocks
• Abstractions
• Mechanisms
• Communication, configuration

16
Technical achievements so far

• Contiki 2.0
• Rime communication stack
• 4668 downloads
• MSPsim MSP430 emulator
• Cooja sensor network simulator
• Sensor node energy monitoring
• Energy profiling for free
• Paper at EmNets IV
• Demos at SenSys, ITEA symposium, today
• Rime/Chameleon
• Makes communication programming easier
• Leverage link-layer mechanisms
• Paper at SenSys

17
We have taken the first steps towards this
Application 2
Application 1
Rime
Best-effort multi-hop unicast
Reliable bulk download
Best-effort network dissemination
Reliable bulk dissemination
Reliable flooding
Reliable data collection
Chameleon
802.15.4
X-MAC
6lowpan IPv6
Wireless HART
18
Clear plan forward

• Holistic system optimization
• Separation of system mechanisms from system
  configuration
• Automatic performance tuning
• Performance monitoring
• Robust reprogramming
• Multi-level watchdog
• System evaluation
• Application trials
• Comparison with the pre-ESNA world

19
Deliverables

• D1.1 State of the art D November 2007
• D1.2 ESNA architecture D November 2007
• D1.3 Contiki 2.0 S April 2007
• D1.4 Simulation platform v1 S February 2008
• D1.5 Contiki 2.1 S February 2008
• D1.6 Guidelines v1 D March 2008
• D1.7 Contiki 2.2 S September 2008
• D1.8 Simulation platform v2 S September 2008
• D1.9 Guidelines v2 D November 2008
• D1.10 ESNA architecture specification
  evaluation
• D November 2008

20
Conclusions

• Highly relevant problem
• Timely
• Unique strengths in ESNA
• Pioneering results already
• Clear plan forward

21
References

• 1 Adam Dunkels, Fredrik Österlind, and Zhitao
  He. An adaptive communication architecture for
  wireless sensor networks. In Proceedings of the
  Fifth ACM Conference on Networked Embedded Sensor
  Systems (SenSys 2007), Sydney, Australia,
  November 2007.
• 2 Adam Dunkels, Fredrik Österlind, Nicolas
  Tsiftes, and Zhitao He. Software-based on-line
  energy estimation for sensor nodes. In
  Proceedings of the Fourth Workshop on Embedded
  Networked Sensors (Emnets IV), Cork, Ireland,
  June 2007.
• 3 Adam Dunkels, Fredrik Österlind, Nicolas
  Tsiftes, and Zhitao He. Demo abstract
  Software-based sensor node energy estimation. In
  Proceedings of the Fifth ACM Conference on
  Networked Embedded Sensor Systems (SenSys 2007),
  Sydney, Australia, November 2007.