Efficient Bio-Sensing Amplifier Design: A Python Based g_m/I_D Design Methodology

Continuous bio-signal sensing applications require circuits with low power consumption and small die area. This creates a need for optimally designed sensing amplifiers balancing noise, power consumption and circuit area. Capacitively coupled instrumentation amplifiers (CCIAs) with chopping and amplifiers biased in weak inversion are a popular design choice for realising low noise amplifiers (LNAs) for bio-signal amplification. In this design overview paper we introduce our open-source Python based $g_{m} / I_{D}$ design tool that enables the fast realisation of optimised bio-signal LNAs. The design uses Jupyter Notebook, facilitating accessible, rapid design and trade-off analysis. A design methodology for realising low noise CCIAs is presented. The trade-off between $g_{m} / I_{D}$ and input-referred noise (IRN) is explored, highlighting the effect of large device sizes in weak-inversion. Trade-offs between circuit area and power consumption for area constrained bio-sensor circuits, especially in the neural sensing domain, are presented. To demonstrate the efficacy of the design methodology a ultralow noise LNA has been designed using a 65 nm technology. The designed circuit is presented with measured chip results demonstrating $2.07 \mathrm{nV} / \sqrt{\mathrm{Hz}}$ in-band noise.