Title :
Profiling energy consumption of Wireless Sensor Nodes with almost zero effort
Author :
Hahm, Oliver ; Adler, Stephan
Author_Institution :
Inst. of Comput. Sci., Freie Univ. Berlin, Berlin, Germany
Abstract :
We present a technique to profile the energy consumption of real Wireless Sensor Network (WSN) applications. The introduced method is highly integrated into the system and allows the developer of WSN algorithms to analyze the power consumption almost without any additional effort. It can break the power consumption down to any desired level from function-based or thread-based analysis to the evaluation of functional blocks. The required measuring setup relies on a simple circuit using an instrumentation amplifier and a shunt. Energy profiling uses features of the GNU Compiler Collection and benefits from the multi-threading operating system μkleos written in pure C code. Compared to similar approaches in this field the energy profiler does not require any modifications to device drivers or application code. Due to this easily deployable method it is feasible to examine the demand for energy for every single node in the network.
Keywords :
device drivers; instrumentation amplifiers; multi-threading; operating systems (computers); program compilers; telecommunication computing; wireless sensor networks; μkleos multithreading operating system; C-code; GNU compiler collection; WSN algorithms; application code; device drivers; energy consumption profiling; function-based analysis; instrumentation amplifier; power consumption analysis; thread-based analysis; wireless sensor network; wireless sensor nodes; Energy consumption; Energy measurement; Hardware; Instruments; Operating systems; Power demand; Wireless sensor networks;
Conference_Titel :
Communications (ICC), 2012 IEEE International Conference on
Conference_Location :
Ottawa, ON
Print_ISBN :
978-1-4577-2052-9
Electronic_ISBN :
1550-3607
DOI :
10.1109/ICC.2012.6364831