http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Maximization of Crossbar Array Memory Using Fundamental Memristor Theory
Eshraghian, Jason K.,Cho, Kyoung-Rok,Iu, Herbert H. C.,Fernando, Tyrone,Iannella, Nicolangelo,Kang, Sung-Mo,Eshraghian, Kamran IEEE 2017 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS PART 2 E Vol.64 No.12
<P>The packing density associated with crossbar arrays offers important architectural solutions to numerous forms of computational engines. Mitigation of sneak paths in the crossbar array, however, requires additional layers in fabrication technology to impede current flow in order to avoid undesired changes to the state when reading and writing to and from the array. This results in an unavoidable increase in the vertical stacking dimension of the array. With the recent emergence of bistable memristors under both dc and ac, by adopting their asymptotic dynamics, we realize a significant improvement in memory construct and spatial constraints of memristor crossbar arrays. In this brief, we formalize a method of configuring a whole array architecture to any permutation of states without sacrificing array density by using a rigorous theoretical analysis, and confirmed via simulation.</P>
Signal Flow Platform for Mapping and Simulation of Vertebrate Retina for Sensor Systems
Cho, Kyoungrok,Baek, Seungbum,Cho, Sung-Wan,Kim, Jun-Ho,Goo, Yong Sook,Eshraghian, Jason K.,Iannella, Nicolangelo,Eshraghian, Kamran IEEE 2016 IEEE SENSORS JOURNAL Vol.16 No.15
<P>Our visual processing system is remarkably good; the retina is nothing like the CMOS image sensor, or for that matter, any of the vision processing architectures that have driven vision systems research for over three decades. Therefore, before embarking upon the complex task of architectural mapping of the retina into hardware, it is essential to gain a realistic insight into the theoretical functions of the retina. In addition, an understanding of the kinds of chemical/electrical interactions taking place must be ascertained in order to venture into the next insurmountable task - the simulation platform. This paper presents a generic signal flow architecture for the mapping of the vertebrate retina derived from ionic current movements and interactions. The approach pursued is focused on the functional behavior of the signal that traverses from the photoreceptor to the ganglion cell in the architecture through transforming the system of nonlinear ordinary differential equations (ODEs) into an equivalent set of non-linear integral equations to cope with the singularity characteristic of retinal systems, providing an increase in the computational speed of similar to 36% when compared with the conventional ODE methods, thus enhancing the realization of a functional retina as part of future hardware-based sensor systems.</P>