Dissertation > Excellent graduate degree dissertation topics show

Based on Genetic Algorithm Approach to build self-feedback test

Author: XiaoZuo
Tutor: ZuoJiShun
School: Hunan University
Course: Information and Communication Engineering
Keywords: Since the feedback test Test Generation Genetic Algorithms Weighted test Reseeding
CLC: TN407
Type: Master's thesis
Year: 2009
Downloads: 55
Quote: 6
Read: Download Dissertation

Abstract

As VLSI technology enters sub-micron and deep sub-micron era, circuits become increasingly complex, increasingly higher degree of integration , which led to the complexity of digital integrated circuits increased sharply . At the same time , people on the reliability of digital integrated circuits are increasingly high requirements , testing of digital integrated circuits technology also will be a considerable development, and to ensure the quality of their products become the decisive factor. This paper reviews the current situation at home and abroad after testing digital integrated circuits , focusing on the self-feedback test generation techniques . The main work includes: 1 . Detailed discussion of the digital IC testing challenges , and describes the most commonly used design for testability techniques . Meanwhile, the analysis of the advantages and limitations of these techniques . Then focuses on several commonly used test generation algorithms. (2) describes the principles and operation of the genetic algorithm , the characteristics of genetic algorithm are analyzed and studied and elaborated genetic algorithm test generation application . In this paper, the self-feedback test generation , the ultimate goal is to search for the best combination of feedback . This adaptive genetic algorithm to search , in a short period of time helps to search the best combination of feedback . 3 This paper studies the self-feedback test generation of key technologies , and proposes a set of tests into the circuit, the internal nodes adjacent vectors response changes caused by the sum of a set of tests as a measure of quality standards . 4 due to a combination of the solution space of great feedback , you need to take some measures to reduce the search range . This article will set multi- weighted method is applied to self- test feedback test generation to limit the weight range of optional internal nodes . Through the benchmark 85 benchmark circuits experiments prove that the method is guaranteed under the premise of fault coverage of the test sequence length is smaller than the similar method has certain advantages. 5 This article is also self-feedback test generation method introduced reseeding . Having identified a set of feedback node , through the feedback node in the group re- seeding to improve fault coverage and shorten the test length , weight of seeds for planting elected by the genetic algorithm . By reference circuit benchmark 85 Experimental results show that the method does not increase the seed in the case , reducing the length of the test sequence .

Related Dissertations

CLC: > Industrial Technology > Radio electronics, telecommunications technology > Microelectronics, integrated circuit (IC) > General issues > Testing and inspection
© 2012 www.DissertationTopic.Net Mobile