Because resistive sensing consumes significant power, in contrast to capacitive sensing, such MEMS have been excluded from all applications whose power supply is restricted. Capacitive transduction is then preferred for many MEMS despite the complex fabrication process and the price increase of the overall application.

LIRMM's team leader Pascal Nouet claimed that resistive sensors can be "a very competitive transduction method if and only if an efficient conditioning and read-out electronics is used."

LIRMM said its proposal consists of a low power interface circuit for resistive sensors. The principle of the conditioning and read-out circuitry, researchers explained, lies in the use of a single current for both resistance biasing and signal amplification. This is achieved by using a common gate MOS configuration of a transistor. If T1 is saturated, it amplifies a small variation of the sensing resistance through a modulation of the gate-source voltage. This means that, with the proposed active bridge, an additional amplifier is no longer required, researchers underlined.

"Assuming the transistor is sized so that it does not degrade the noise level, we may expect from the active bridge the same SNR and the same power consumption as for the bare sensor but an output voltage at least two orders of magnitude higher," Nouet explained.

The architecture can be used either to improve the SNR of a resistive sensor or to reduce the overall power consumption and the complexity of the interface. Simulation and silicon results, demonstrate a current consumption divided by ten, or a factor three SNR improvement, LIRMM claimed.

This patented technology is available for in-licensing.