International Journal of Grid and Distributed Computing Vol. 11, No. 3 (2018), pp.51-62 http://dx.doi.org/10.14257/ijgdc.2018.11.3.05 ISSN: 2005-4262 IJGDC Copyright ⓒ 2018 SERSC Australia Impact-based Piezoelectric Energy Harvester as a Power Source for a Neural Activity Monitoring Circuit * Sunhee Kim 1* , Suna Ju 2 and Chang-Hyeon Ji 3 1 Department of System Semiconductor Engineering, Sangmyung University 2,3 Department of Electronic and Electrical Engineering, Ewha Womans University 1 happyshkim@smu.ac.kr Abstract Recently, energy harvesters have been studied as a power source for implantable biomedical devices. However, practical applications of energy harvesters to implantable medical devices are very rare because energy harvesters and implantable medical devices have been studied separately. In this paper, we present a neural activity monitor circuit, which is one of implantable medical devices, solely powered by an energy harvester. The system consists of an impact-based piezoelectric energy harvester, a power transfer circuit, and a neural signal monitoring circuit. Reliability of the system was verified by comparing it with a system powered by a DC power supply. It was found that the harvester provided sufficient energy to the system for reliable neural activity monitoring. We expect that presented results may work as a reference for designing harvesters and power transfer circuits together with implantable medical devices in an environment of a human body. Keywords: biomedical monitoring, energy harvesting, equivalent circuits, vibration energy 1. Introduction As portable devices, such as wireless sensors, personal information devices and implantable medical devices, have been widely utilized, their power source has become increasingly important. Because they have usually used batteries as their power source and batteries have a limited life, dead batteries must be replaced after a certain period of time. In the case of wireless sensors, using disposable batteries limits service locations and prevents a sensor network from scaling to hundreds or thousands of nodes [1]. In the case of implantable biomedical devices, such as pacemakers [2], electrical stimulators [3], and closed-loop seizure controllers [4, 5], battery replacement can cause physical pain, psychological distress, and economic burden for a patient [6, 7]. Recently, energy harvesters have attracted attention because they are expected to replace batteries [8-10]. Energy harvesting techniques have been studied to convert various energy sources, such as solar, thermal energy, acoustic energy, and kinetic energy into electrical energy [11]. Solar energy harvesters have been usually used in wireless sensor network (WSN) because they provide relatively high power density and voltage/current levels required for WSN devices [12]. Thermal energy, bio fuel cells and kinetic energy from body motion have been studied in implantable biomedical fields [6, 13]. Harvesters based on kinetic energy generated by human-body motion are considered as be more useful in powering implantable devices at this time [6, 13, 14]. Human-motion induced vibration can usually be characterized by relatively low frequency and the random nature [15]. Power Standards Lab shows resonance frequency of human body ranges from 4-5 Hz to 50-100 Hz Received (December 19, 2017), Review Result (February 5, 2018), Accepted (February 7, 2018) * Corresponding Author