A Highly Linear Analog-to-Spike Converter
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
This brief presents a low-voltage differential voltage-to-spike encoder based on a bulk-driven transconductor and dual leaky integrate-and-fire (LIF) neurons. The proposed transconductor employs a bulk-driven differential input pair with diode-connected source degeneration to enhance linearity under reduced supply voltage operation. The circuit converts an input differential voltage into complementary output currents, which are directly injected into two LIF neurons to encode the input amplitude into spike firing rates. The two neurons generate opposite-polarity spike trains (0–VDD and 0–VSS), enabling fully differential spike encoding with improved dynamic range, common-mode rejection, and noise robustness compared to single-ended encoders. The design was implemented and simulated in a standard 0.18-μm CMOS technology, demonstrating low-voltage operation, enhanced linearity, and wide input dynamic range suitable for neuromorphic and mixed-signal sensor interface applications.