Konu "Algorithm design and analysis" ile Elektrik Elektronik Fakültesi'a göz atma
Sayfa başına sonuç
ÖgePermanent and transient fault tolerance for reconfigurable nano-crossbar arrays( 2016-08-25) Tunali, Onur ; Altun, Mustafa ; https://orcid.org/0000-0002-3103-1809 ; Department of Nanoscience and Nanoengineering ; Department of Electronics and Communication EngineeringThis paper studies fault tolerance in switching reconfigurable nano-crossbar arrays. Both permanent and transient faults are taken into account by independently assigning stuck-open and stuck-closed fault probabilities into crosspoints. In the presence of permanent faults, a fast and accurate heuristic algorithm is proposed that uses the techniques of index sorting, backtracking, and row matching. The algorithm's effectiveness is demonstrated on standard benchmark circuits in terms of runtime, success rate, and accuracy. In the presence of transient faults, tolerance analysis is performed by formally and recursively determining tolerable fault positions. In this way, we are able to specify fault tolerance performances of nano-crossbars without relying on randomly generated faults that is relatively costly regarding that the number of fault distributions in a crossbar grows exponentially with the crossbar size.
ÖgeYield analysis of nano-crossbar arrays for uniform and clustered defect distributions(IEEE, 2018-02-15) Tunali, Onur ; Altun, Mustafa ; https://orcid.org/0000-0002-3103-1809 ; Electronics and Communication EngineeringDuring the fabrication of nano-crossbar arrays, certain amount of defective elements are introduced to the end product which affect the yield drastically. Current literature regarding the yield analysis of nano-crossbar arrays is very rough and limited to the uniform distribution of defect occurrence with a few exceptions. Since density feature of crossbar architectures is the main attracting point, we perform a detailed yield analysis by considering both uniform and non-uniform defect distributions. Firstly, we briefly explain the present algorithms and their features used in defect tolerant logic mapping. Secondly, we explain different defect distributions and logic function assumptions used in the literature. Thirdly, we formalize an approximate successful mapping probability metric for uniform distributions and determine area overheads. After that, we apply a regional defect density analysis by comparing uniform and clustered defects to formulate a looser upper bound for area overheads regarding clustered distributions. Finally, we conduct extensive experimental simulations with different defect distributions.