To get the smaller size, the ARM Cortex-M0+ microcontroller core and temperature sensor only retain data when powered by an LED source to a photovoltaic cell. This source is also used to transmit and receive data.
Designed as a precision temperature sensor, the new device converts temperatures into time intervals, defined with electronic pulses. The intervals are measured on-chip against a steady time interval sent by the basestation that provides power via an LED, and the data then converted into a temperature. As a result, the sensor can report temperatures in tiny regions such as a cluster of cells with an error of about 0.1ºC.
The team had already developed the Michigan Micro Mote which measures 2x2x4mm and retains its state when there is no power, leading to a debate on what constitutes a computer system. Like this new design, the IBM system does not retain data unless powered.
“We are not sure if they should be called computers or not. It’s more of a matter of opinion whether they have the minimum functionality required,” said David Blaauw, a professor of electrical and computer engineering (ECE), who led the development of the new system together with ECE professors Dennis Sylvester and Jamie Phillips.
One of the big challenges in making the system was how to run at very low power when the system packaging had to be transparent. The light from the base station—and from the device’s own transmission LED—can induce currents in the circuits. This meant using switched capacitors instead of diodes across the design, and the team worked with Mie Fujitsu Semiconductor Japan and Fujitsu Electronics America on the implementation.
“We basically had to invent new ways of approaching circuit design that would be equally low power but could also tolerate light,” said Blaauw.
The team chose precision temperature measurements because of a need in oncology. A longstanding collaborator, Gary Luker, a professor of radiology and biomedical engineering, wanted to answer questions about temperature in tumors. Some studies suggest that tumors run hotter than normal tissue, but the data isn’t solid enough for confidence on the issue. Temperature may also help in evaluating cancer treatments.
“Since the temperature sensor is small and biocompatible, we can implant it into a mouse and cancer cells grow around it,” said Luker. “We are using this temperature sensor to investigate variations in temperature within a tumor versus normal tissue and if we can use changes in temperature to determine success or failure of therapy.”
The system could be used for a wide range of applications, says Blaauw. “When we first made our millimeter system, we actually didn’t know exactly all the things it would be useful for. But once we published it, we started receiving dozens and dozens and dozens of inquiries,” he said. This could include pressure sensing inside the eye for glaucoma diagnosis, oil reservoir monitoring, biochemical process monitoring and audio and visual surveillance.
The paper titled “A 0.04mm3 16nW Wireless and Batteryless Sensor System with Integrated Cortex-M0+ Processor and Optical Communication for Cellular Temperature Measurement” was presented at the 2018 Symposia on VLSI Technology and Circuits.