한양대학교 ERICA 산학협력단

> 연구정보 > 연구성과 > 금주의 우수논문

한양대학교 재직중인 전임교원이 출판한 SCIE급 논문중 상위 10%이내 논문 (인문계열 SSCI 상위50%이내)
매주 업데이트

SCI Article

Spectral Leakage-Driven Loopback Scheme for Prediction of Mixed-Signal Circuit Specifications

	성명	김병호 ()
	소속	공학대학 전자공학부
	캠퍼스
	우수선정주	2019년 05월 3째주

Author

김병호 (Dept Elect Engn) corresponding author; Jacob A. Abraham;

Corresponding Author Info

Kim, B (reprint author), Hanyang Univ, Dept Elect Engn, Ansan 15588, South Korea.

E-mail

brandonkim@hanyang.ac.kr

Document Type

Article

Source

IEEE Transactions on Industrial Electronics Volume:66 Issue:1 Pages:586-594 Published:2019

Times Cited

External Information

https://ieeexplore.ieee.org/document/8345712

Abstract

The rising cost of production testing for a system-on-a-chip (SoC) is one of the crucial matters to chip makers, due to long test time and costly automated-test-equipment. This paper proposes a spectral leakage-driven built-in self-test (BIST) scheme to precisely predict the nonlinearity of mixed-signal circuits in the loopback mode, thereby accomplishing cost-effectiveness (compared to previous BIST-based works). A digitally synthesized single-tone sinusoidal stimulus used for conventional harmonic testing is incoherently sampled by a device under test (DUT). The DUT output signal exhibits the correlation between the DUT harmonics and the spectral leakage introduced by the incoherent sampling. The DUT output signal is then fed to another DUT through a loopback path, so that the harmonics of a pair of DUTs are correlated with the spectral leakage on the loopback response; the magnitude of the spectral leakage is considered as a weighting factor on the harmonic magnitude of those DUTs. The correlation is quantitatively modeled as characteristic equations in (15), and postprocessing predicts the harmonics of the two individual DUTs, by simultaneously solving the characteristic equations using on-chip DSP core available in an SoC. Simulation and hardware measurements validated that this paper can be practically used for production testing by showing less than 0.3 and 0.6 dB of the prediction errors, respectively.

Web of Science Categories

Engineering

Funding

Basic Science Research Program through the National Research Foundation of Korea - Ministry of Education [2016R1D1A1B02013884]

Language

English

attached file