Energy Scavenging in Support of Ambient Intelligence: Techniques, Challenges, and Future Directions

Abstract

Ambient intelligence (AmI) depends on the existence of vast quantities of wireless sensors distributed throughout the environment. While advances in IC fabrica- tion technologies, circuit designs, and networking techniques have greatly reduced the cost, size, and power consumption of potential wireless sensor platforms, the development of suitable power sources for many applications lags. The purpose of this chapter is both to review technologies for scavenging energy from the environment to power wireless sensors, and to discuss challenges and future research directions for vibration-based energy scaven- ging. Many potential energy scavenging technologies are presented along with current state of research and theoretical maximum power densities. Special focus is given to scavenging energy from mechanical vibrations extant in many environments. Results from vibration- based energy scavengers using piezoelectric structures developed by the authors are pre- sented demonstrating power production of approximately 375 mW=cm 3 . While wireless sensor nodes have successfully been powered by vibration-based energy scavengers, several improvements are possible, and indeed necessary, for widespread deployment. Three areas