利用平行測量與模型化隨機森林預測臨界電壓平均數與變異數

Abstract

在傳統晶圓接受度測試的測試結構下，測量元件臨界電壓之變異程度需要很長的測試時間。本論文提出一個模型擬合程序，可以快速且有效地獲得大量待測元件臨界電壓之平均數與變異數。所提出的程序採用模型化隨機森林作為其核心的模型擬合技術，僅基於待測元件並聯連接之閘極總電流即可擬合出足以預測臨界電壓平均數與變異數之預測模型。根據聯電28奈米製程技術之積體電路仿真模擬程式進行實驗，研究結果顯示，所提出的擬合模型對於臨界電壓平均數與變異數之預測均可達到超過99%決定係數。相較於使用二元搜尋法的傳統晶圓接受度測試，若以平均每顆待測元件所需的閘極電流測量次數，我們所提出的程序可以達到42.9倍的增速。

To measure the variation of device Vt requires long test for conventional WAT test structures. This thesis presents a model-fitting framework that can efficiently and effectively obtain the mean and variance of Vt for a large number of DUTs. The proposed framework applies the model-based random forest as its core model-fitting technique to learn a model that can predict the mean and variance of Vt based on only the combined Id measured from parallel connected DUTs. The experimental results based on the SPICE simulation of a UMC 28nm technology demonstrate that the proposed model-fitting framework can achieve a more than 99% R-squared for predicting both of Vt mean and variance. Compared to conventional WAT test structures using binary search, our proposed framework can achieve 42.9X speedup in turn of the required iterations of Id measurement per DUT.