Seminar: Long-running computation under transient power conditions using FRAM
Speaker: Domenico Balsamo
Time: 28/11/2013 16:00-17:00
Abstract:Transiently powered embedded systems such as Wireless Sensor Networks (WSNs) with Energy Harvesting (EH) solutions typically rely on programs that complete a task under tight time constraints before energy starvation leads to complete loss of volatile memory. With ultra-low power Microcontrollers (MCUs), such as the MSP430 microcontroller family, the WSNs can perform computation and sensing under tight energy conditions. However, these MCUs consume more power than conventional energy harvesting circuitry, and energy consumption can easily outpace harvesting, resulting in frequent power loss. Computing under transient power conditions is a challenge. The main idea of my work is to take advantage of FRAM memory, a new non-volatile storage technology that allows lower power usage, faster write performance and a greater number of write-erase cycles, to enable long-running computations to span power loss events. For these reasons, the FRAM memory will be useful as it allows the software to remember its state through such disruptions. As a result, we can present a brand new paradigm for EH-WSNs, whereby they can operate directly from the energy source. This work is focused on implementing run-time software, for the MSP430FR57xx family of microcontrollers, which allows interrupting the execution of the main program when the voltage goes below a given threshold. It is therefore possible to predict power losses and to save the entire state of the system into non-volatile FRAM memory. After a failure, program state is restored and the execution continues rather than restarting from the beginning.
Biography: Domenico Balsamo received a Master’s Degree in Electronic Engineering from the University of Modena and Reggio Emilia, Italy, in 2008. From 2009 to 2011 he was with Conchiglia S.p.a. as Project Manager in AC/DC and DC/AC power converters for the reduction of the irradiance in public illumination equipment. Since January 2012, he is pursuing his Ph.D. in Electrical and Computer Engineering Department at the University of Bologna, Italy. His current research interest is focused on HW/SW design of Wireless Sensor Networks (WSNs) for monitoring the energy efficiency of smart buildings, under the supervision of Prof. Luca Benini. His research is supported by Telecom Italia S.p.a. and covers the design and optimization of low-power and low-cost WSNs for climate and environmental monitoring, the exploration of new techniques for Energy Scavenging, the development of new algorithms for WSNs communication, and new standards exploration.
Posted by Dr Geoff Merrett on 30 Oct 2013.