Full metadata record
DC FieldValueLanguage
dc.contributor.author郭淳仁en_US
dc.contributor.authorKuo, Chun-Renen_US
dc.contributor.author趙家佐en_US
dc.contributor.authorChao, Chia-Tsoen_US
dc.date.accessioned2014-12-12T01:37:41Z-
dc.date.available2014-12-12T01:37:41Z-
dc.date.issued2010en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#GT079711671en_US
dc.identifier.urihttp://hdl.handle.net/11536/44370-
dc.description.abstract這篇論文提出掃描鏈的重新排列架構在低功率固定錯誤樣式填充的基礎上,增加橋接錯誤故障涵蓋率。為了克服功率固定錯誤樣式填充所帶來的困難,我們先從物理佈局上面抽出可能的橋接錯誤的發生點,接著利用這些發生點計算每一對PPI在橋接錯誤的可能觸發數,並用這個數字當作每一對PPI的影響力。接著我們提出掃描鏈的劃分以及掃描鏈的重新排列來減少每一條掃描鏈內所有掃描相對間的關聯性,進而提高橋接錯誤故障涵蓋率。最後我們做一些實驗來驗證方法的有效性,繞線的影響也投過實驗來做討論。zh_TW
dc.description.abstractThis thesis proposes a scan reordering method based on low-power fill on stuck-at fault pattern to improve the bridge fault coverage. In order to overcome the drawback of low-power fill on bridge fault coverage, we first extract bridge faults from physical layout and calculate the impact of each PPI pair on bridge-fault activation to measure the correlation between two PPI pair. Next, scan partition and scan reorder are performed to generate scan order that minimizes the correlation hold by scan cells in each scan chain. Experiment shows our proposed method generates better bridge fault coverage comparing to random scan order and the wire length overhead is also provided. Finally, some future possible improvement is presented.en_US
dc.language.isozh_TWen_US
dc.subject掃描鏈zh_TW
dc.subject掃描鏈重新排序zh_TW
dc.subject橋接錯誤zh_TW
dc.subject測試zh_TW
dc.subjectTestingen_US
dc.subjectScan reorderen_US
dc.subjectbridge faulten_US
dc.subjectlow-power pattern fillen_US
dc.title利用掃描鏈重新排序增加橋接錯誤故障涵蓋率架構在低功率固定錯誤樣式填充zh_TW
dc.titleScan reordering for improving bridge fault coverage based on low-power stuck-at-fault pattern fillen_US
dc.typeThesisen_US
dc.contributor.department電子研究所zh_TW
Appears in Collections:Thesis


Files in This Item:

  1. 167101.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.