Research

Background: When an ADC is converting a confidential analog signal, it may leak side-channel information that can be used to recover the secret signal. Preventing an ADC from leaking such secrets on silicon is an active research topic. We are interested in:

1. How to benchmark side-channel security and the security-overhead tradeoff for ADCs?

2. More secure ADC designs with less power, performance, and area (PPA) overhead

Results: [OJCAS2025] [MDTS2026] We proposed and experimentally verified a proxy ADC framework that mathematically relates signal-to-noise (SNR) to normalized root-mean-square error (NMSE). We also argue that the random number generator itself should be included in the hardware security cost and overhead.

Why it matters: If an attacker can infer whether you are at home by hacking your energy meter, would you be worried?   

Background: Low Earth Orbit (LEO) and deep space present a very harsh environment for microelectronics to operate due to the "short-term" Single-Event Effects (SEE) and "long-term" Total Ionizing Dosage (TID).  Our work seeks to Radiation Harden Analog / Mixed-Signal circuit building blocks By Design (RHBD) instead of relying on specialized processes.

The Bindra comparator [1] is known for its energy efficiency. When used in a SAR ADC, it achieved the lowest Walden Figure of Merit [2, 3]. However, the reliability of such a comparator in space applications is unknown. This study examined the potential of adapting the best-performing, energy-efficient terrestrial ADCs for space applications and ways to mitigate their single-event vulnerabilities. 

Results: [IEEE Xplore] [MDTS2025] 

Why it matters: Space science equipment needs highly reliable ADCs to measure particles, among others.  

References: [1] Bindra18JSSC, [2] Bindra17VLSI, [3] ADC survey.

Sponsor: NASA Oklahoma EPSCoR office, Research Initiation Grant (RIG)