Personal tools

Flexible Front-End Circuit for Biomedical Data Acquisition

From iis-projects

Revision as of 10:53, 10 March 2015 by Weberbe (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
Chip micrograph of the 8-channel sensor front-end and data acquisition IC ‘Cerebro v2’ for electrode-based medical applications, such as ECG or EEG. Implemented in a 130 nm CMOS technology.


2010 - 2013


Philipp Schönle
Schekeb Fateh


Nano-Tera: PlaCITUS




Biomedical signals measured for ECG, EEG, and EMG are electrical potentials caused by neural activity. These signals are typically picked-up by electrodes which are placed on the patient’s skin. The electrode-skin contact is a metal-to-electrolyte interface, resulting in the superposition of DC half-cell potentials to the signal of interest. This offset voltage can exceed the signal amplitude (100 μV ...1 mV) by three orders of magnitude and is varying over time since it is affected by the electrolyte, the skin’s condition and the contact pressure.

The ASIC developed in this project consists of an 8 channel, low-noise, high input-impedance, and programmable-gain analog front-end and an ADC shared by all eight channels. To avoid saturation by the above mentioned high offset, the latter is subtracted from the signal with a current-steering DAC prior to amplification. While this eight channels are optimized for low-amplitude differential signals, an auxiliary (AUX) ADC included in this chip allows to measure single-ended input signals of up to 2.4 V or alternatively the temperature by an integrated sensor.

The differential channels have an input-referred noise of 82 nV/√Hz, a maximum sampling rate of 8 kS/s, a configurable gain between 128 and 2048, and tolerate a maximum signal offset of ±300 mV.