Technology

Smart microchip operates even if battery runs out

Internet of Things (IoT), though still in infancy, shapes the future of many industries and will also affect everyday life in many and important ways. One of the key challenges of creating efficient IoT devices from the idea is in fact long-term operation with strictly limited power sources and therefore extreme power efficiency.

IoT devices are deployed on a large scale with built-in sensors, which differentiates them from other devices, additionally some of the organizations, services and users that use them are usually located in remote locations away from urban centers, so autonomy is a necessary feature to make a commercially viable solution for the general consumer. At present, batteries for IoT devices are much larger than the chip that feeds, and up to three times more expensive than that. The size usually determines the life of the sensor, which directly affects how often it needs to be changed.

This greatly affects maintenance costs and environmental impact, all because of batteries. To extend the overall life, the battery is usually slowly recharged by collecting some limited power from the environment, such as solar energy. However, existing IoT devices can not operate without battery and small batteries are often discharged completely. Therefore, battery disadvantages often result in extremely inconsistent operation of IoT devices, as they stop working whenever the battery runs out. To address this technological divide, a team of engineers from the National University of Singapore (NUS) has developed an innovative microchip called BATLESS, which can continue to function even when the battery runs out.

BATLESS is designed with a new power management technique that allows it to start and operate with low light without battery assistance, using a very small solar cell on the chip. This discovery of research greatly reduces the size of the batteries needed to feed the IoT sensor nodes, making them 10 times smaller and cheaper to produce. The revolution has been presented at the conference of the International Conference on Solid State Circuits (ISSCC) 2018 in San Francisco, the world’s leading forum for presenting advances in solid state circuits and on-a-chip systems.

The head of the NUS research team, Associate Professor Massimo Alioto of the NUS Engineering School, said: “We have proven that batteries used for IoT devices can shrink considerably, as they do not always have to be available to maintain continuous operation their.  Addressing this fundamental problem is a major advancement towards the ultimate uninterrupted use of IoT sensors without the need for batteries, and it is expected to pave the way for a world of trillions of IoT devices around the globe. The ability of IoT devices to continue operating even when the battery runs out, practically changes the data so far and is considered a great breakthrough.

This process is accomplished in two ways – minimum energy and minimal power. When battery power is available, the chip operates in a minimum power state to maximize battery life. However, when the battery runs out, the chip switches to minimal power mode and operates at a very low power consumption of about half a nano-watt, that is, it is about one billion times lower than the power consumption of a smartphone during a phone call. Power can be provided by a very small solar cell on the chip that is about half a millimeter in size, In addition, energy can be collected in other forms of environment, such as vibration and heat.

The ability of the microprocessor to change the state of operation between minimum power and minimal energy translates into aggressive reduction of batteries from a hundred to several millimeters. The BATLESS Microprocessor allows the unusual detection, processing, recording and timing of events of interest and the transfer of such valuable data to the cloud when the battery becomes active again. Although when in minimum power mode when the battery will not be active, the reduced microchip speed will still be sufficient for many IoT applications needed to detect parameters that vary in