A Standard-Cell-Based Neuro-Inspired Integrate-and-Fire Analog-to-Time Converter for Biological and Low-Frequency...

Area of Science:

• Electrical Engineering
• Signal Processing
• Biomedical Engineering

Background:

• Continuous-time asynchronous data converters, like analog-to-digital and analog-to-time converters, are crucial for processing sparse biological signals.
• The integrate-and-fire converter (IFC), inspired by neural systems, offers a potential solution for such applications.

Purpose of the Study:

• To design and prototype two novel IFC circuits using a 130 nm CMOS standard process: a standard-cell-based (SCB) open-loop dynamic IFC and a closed-loop analog IFC.
• To conduct a comparative analysis of the performance characteristics of the two proposed IFC designs.

Main Methods:

• Prototyping of two distinct IFC circuits: a novel SCB open-loop dynamic IFC and a closed-loop analog IFC.
• Utilizing a 130 nm CMOS standard process for circuit fabrication.
• Comparative analysis of power dissipation, energy per pulse, signal amplitude and frequency ranges, conversion errors, and maximum pulse density.

Main Results:

• The SCB IFC achieved a power dissipation of 59 μW and an energy per pulse of 18 pJ, significantly lower than the analog IFC's 53 μW and 1060 pJ.
• Both converters operated without an external clock, handling signal amplitudes from 1.6 mV to 28 mV (SCB) and 0.6 mV to 2.4 mV (analog), with low normalized RMS errors (<5.2%).
• The SCB IFC exhibited a maximum pulse density of 3300 kHz, substantially higher than the analog IFC's 50 kHz.

Conclusions:

• The SCB IFC offers one of the lowest energy per pulse consumptions reported for IFC circuits, highlighting its efficiency.
• The fully differential analog IFC represents a novel contribution to the field of IFC design.
• Both developed IFC circuits demonstrate promising performance for applications requiring efficient, clock-less data conversion, particularly the SCB variant for its low energy consumption and high pulse density.