Exploitingperiodic motion insensor networks wireless sensor networks (Masterthesis)


Jing Han

30.04.2014 13:00, room 3945


In recent years, there has been an increasing research interest in
reducing energy consumption for all body area sensor networks, since
the need for a longer battery run-time on these devices is increasing.
To design sensor networks with lower energy consumption, methods
including different protocols design, optimized duty cycles, and
transmission power level reducing are used. Whereas the transmission
power level is set for reliable transmission for the highest path loss
that might occur within the body area sensor network, which is
redundant for most of time. To reduce the power consumption further,
we suggest a novel mechanism for exploiting periodic motion. Unlike
sending packets from arbitrary locations with a quite high transmission
power level to overcome the high attenuation, our proposed method
predict the locations with lowest path loss and send only at these
points with a reduced transmission power level. Therefore, more energy
is saved. The prediction of these best reception points are based on a
correlation between the acceleration data and the received signal
strength, during these exploitable periodic movements. The performance
of the method was evaluated with six volunteers and 2 cases each volun-
teer and resulted in a power reduction up to 24.7% for the
transmission of a packet while a low packet error rate was still