When Timing Matters

Olaf Landsiedel, Muhammad Hamad Alizai, Klaus Wehrle

Abstract:

The rising complexity of data processing algorithms in sensor networks combined with their severely limited computing power necessitate the in-depth understanding of the timely behavior of implementations. However, today only cycle accurate emulation and test-beds provide a detailed and accurate insight into the timely behavior of sensor networks. We introduce fine grained, automated instrumentation of simulation models with cycle counts derived from sensor nodes and application binaries to provide detailed timing information. The presented approach bridges the gap between scalable but abstracting simulation and cycle accurate emulation for sensor network evaluation. By mapping device specific code with simulation models, we can derive the timing of each source code line on a sensor node. As a result of such a mapping it is possible to determine the time and duration a certain code line takes to get executed. Hence, eliminating the need to use expensive instruction level emulators with limited speed, restricted scalability and portability. Furthermore, the proposed design is not bound to a specific hardware platform, a major advantage compared to existing emulators. Our evaluation shows that the proposed technique achieves a timing accuracy of 99% compared to emulation while adding only a small overhead. Concluding, it combines essential properties like accuracy, speed and scalability on a single simulation platform.

Key Features:

• Time accurate simulation of the application and operating system source-code using automatic code mapping.
• Delaying and interrupting simulation events to accurately model the priority based interrupt properties of the hardware platform.
• Automatic, static and manual mapping of the simulation specific TOSSIM components to reinforce time accuracy.
• Atomic statement modeling to disable interrupts during the execution of atomic block of statements.
  

To Do:

• Further strenghtening of the code mapping algorithm to remove inaccuracies and adding support for different compiler optimizations.
• Fine granular implementation of the CC2420 and CC1000 radio stacks for TimeTOSSIM, as well as profiling radio related components to provide accurate timing in radio communication.
• A plug-in based multi-platform support (telosb, mica2 etc..).
  
• Support for online (during the simulation run) energy modelling.
  

• Olaf Landsiedel, Hamad Alizai, Klaus Wehrle: "When Timing Matters: Enabling Time Accurate & Scalable Simulation of Sensor Network Applicaitons", ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), (Washington U.) St. Louis, Missouri, April 2008
• Muhammad Hamad Alizai, Olaf Landsiedel, Klaus Wehrle: "Accurate Timing in Sensor Network Simulation", 6th GI/ITG KuVS Fachgespraech "Wireless Sensor Networks", Aachen, Germany, July 2007 (ISSN#0935-3232)