An Aussie Science Breakthrough Could Reinvent Wi-Fi

Image: uPanel

The internet of things is built on tiny, low-power, often wireless sensors that have small and very specific tasks. These devices are usually battery powered, which is convenient at low energy usage levels but which can be an impediment to long-term use when those batteries regularly require replacement. A team of researchers at the Australian National University has accurately modeled how much energy the wireless transfer of information takes from low-power sensors, which is the first step in designing devices that can harvest power from the ambient radio frequency communications in the air around them.

The research, done out of Australian National University in Canberra and led by Dr Salman Durrani, is years away from a practical application. But the research, published in the scientific journal IEEE Transactions on Wireless Communication, points to energy harvesting being a feasible method of powering low-energy wireless sensors either through solar power or even the ambient radio signals -- from sources like mobile phone base stations or AM/FM radio braodcasting towers.

Energy harvesting is "inherently random", according to the research abstract, but is promising for powering not only the transmission of data from wireless sensors but also the sensing that those devices undertake, which can be for anything from temperature or humidity for crops to the condition of structures like bridges. Previous research has only covered the energy cost of data transmission to a central node, not the cost of sensing itself -- which is done on current devices through battery power, eventually requiring replacement.

It's overly optimistic to think that energy harvesting will have a practical future for mainstream wireless long-distance charging of consumer devices like smartphones or electric cars, but the ANU research suggests that for the imminent, ubiquitous network of low-power Internet of Things sensors, energy harvesting could be used not only to enable wireless transmissions but also the process of sensing itself. Because there's a trade-off between power usage for sensing or transmission, a compromise is necessary.

With this compromise in mind for the study, ANU researchers found that when wireless sensors were powered only by energy harvesting -- with no battery required -- a device that took a sensor reading and transmitted it (a harvest-then-use or "save-then-transmit" protocol") would only experience a delay of a few hundred milliseconds while energy harvesting took place. This is a long time in in the life of a wireless signal, but still has practical applications.

While there's a long way to go until we see Internet of Things gadgets that are completely powered only by the wireless signals in the air around us, this ANU research is a promising first step.

Energy harvesting (EH) provides a means of greatly enhancing the lifetime of wireless sensor nodes. However, the randomness inherent in the EH process may cause significant delay for performing sensing operations and transmitting sensed information to the sink. Unlike most existing studies on the delay performance of EH sensor networks, where only the energy consumption of transmission is considered, we consider the energy costs of both sensing and transmission. Specifically, we consider an EH sensor that monitors some status property and adopts a harvest-then-use protocol to perform sensing and transmission.    To comprehensively study the delay performance, we consider two complementary metrics and analytically derive their statistics: 1) update age—measuring the time taken from when information is obtained by the sensor to when the sensed information is successfully transmitted to the sink, i.e., how timely the updated information at the sink is, and 2) update cycle—measuring the time duration between two consecutive successful transmissions, i.e., how frequently the information at the sink is updated. Our results show that the consideration of sensing energy cost leads to an important tradeoff between the two metrics: more frequent updates result in less timely information available at the sink.