Speaker: Tao Chen, ECpE Graduate Student
Advisor: Degang Chen
Title: Analog-to-digital Converter Built-in Self-test and Self-healing
Abstract: Linearity test time often dominates analog-to-digital converter’s (ADC) test costs. State-of-the-art linearity tests use histogram methods that require highly linear signal sources and many samples per ADC code. This results in high test complexity, expensive testing platforms, and long test times. An ultrafast and accurate built-in-self-test (BIST) approach for full parametric test and test-based self-calibration of embedded ADCs is proposed. It combines the ultra-fast and accurate linearity test methodology with the nonlinear excitation approach to obtain ultra-fast and accurate linearity test results of ADCs with imprecise signal generator. Two identical nonlinear input signals with constant voltage shift between them are applied to the ADC. The nonlinearity from the input signal is removed with the proposed method thus achieving accurate estimation. The proposed method can be implemented on chip with minimal cost overhead, providing a BIST solution for deeply embedded ADCs. Furthermore, the self-test results from the ADC BIST are used for self-healing to repair the ADC and achieve better performance.