Improving energy efficiency in IoT with re-configurable virtual objects

In this paper we present a method for reconfiguring Virtual Objects (VOs) during run-time. VOs are semantic descriptions of ICT objects and the associated physical objects and phenomena they observe. VOs also include software modules to expose ICT object functionalities as IoT services for re-use. In our case study we show, by using an analytical model, how the energy efficiency of a wireless battery powered weather station (WS) can be improved. We introduce descriptions of the available operating modes and store them in the WS semantic description. These operating modes of the WS´s VO are linked with the ability of the ICT object to change the compression method used to transmit information during run-time. We evaluate the energy efficiency and latency of three operating modes, namely uncompressed, lossy and lossless modes. The results show that the possibility to re-configure the operating mode during run-time from uncompressed to lossy mode lowers the total energy for transmission to 47.9%.