POWER OPTIMIZATION FOR WIRELESS AUTONOMOUS TRANSDUCER SOLUTIONS Valer Pop 1 , Jef van de Molengraft 1 , Francois Schnitzler 2 , Julien Penders 1 , Rob van Schaijk 1 , Ruud Vullers 1 1 Holst Centre / IMEC-NL, Eindhoven, the Netherlands 2 Montefiore Institute, University of Liege, Liege, Belgium Abstract: Low-power and small size are key demands for wireless autonomous transducer solution (WATS) systems. This demand has motivated industry and research institutions to work on various advanced small size energy systems (ES) that can efficiently deliver power to demanding applications. In order to enable low-power and consequently autonomy, power optimization at WATS system level is crucial. This paper deals with innovative power optimization techniques for two different WATS applications that transmit data at high or low duty-cycle, respectively. By applying power optimization techniques significant improvements in power consumption have been obtained. The results show the effectiveness of our low-power design techniques and power optimization approach for improving the WATS power consumption. Subsequently, the limitations of WATS applications integrating ‘off-the-shelf’ energy systems and low-power electronics are also revealed. Key words: low-power, wireless, harvester, energy, autonomous 1. INTRODUCTION This paper discusses the trade-offs between power consumption, duty-cycle and functionality for Wireless Autonomous Transducer Solution (WATS) applications. A WATS system typically contains different building blocks, schematically shown in Figure 1, where a system integration approach is needed for low-power optimization. This includes but is not limited to understanding the energy harvester and energy storage system (ESS) performances, and using ultra low-power technologies for sensing, data processing and wireless communication (see Figure 1). Previous work focused on energy harvesters [1]- [6], batteries [7]-[9] or power management circuits [10] for WATS applications. However, whilst the topics of energy generation and conversion have been the subject of extensive research, system integration and power optimization techniques at system level have not been sufficiently addressed. This work therefore focuses on the development of two new ultra-low power WATS applications fully powered by indoor photovoltaic (PV) energy harvesters. Subsequently, the limitations of WATS applications integrating ‘off-the-shelf’ energy systems and low- power electronics are discussed. Energy storage - battery - supercapacitor AC/DC or DC/DC DC/DC converter Scavenger Battery state Sensor ADC Radio Processor Power Data converter Power management Fig. 1: Schematic of a WATS node with data transfer (dashed line) and power transmission (continuous line). The system input power is delivered by the harvester and energy storage system combination. Typical load modules are the sensor, Analog-to-Digital Converter (ADC), processor and radio. Proceedings of PowerMEMS 2008+ microEMS 2008, Sendai, Japan, November 9-12, (2008) 141