Browsing by Author "Li, Xin"

Now showing 1 - 3 of 3

A Novel Analog Physical Synthesis Methodology Integrating Existent Design Expertise
(2015-02-01)
Wu, Po-Hsun
;
Lin, Mark Po-Hung
;
Chen, Tung-Chieh
;
Yeh, Ching-Feng
;
Li, Xin
Analog layout design has been a manual, time-consuming, and error-prone task for decades. To speed up layout design time for a new design, analog layout designers prefer referring to legacy designs and layouts rather than starting from scratch, or thoroughly applying placement and routing tools because legacy layouts contain pretty much design expertise. Motivated by such layout design process, this paper presents the first knowledge-based physical synthesis methodology to generate new layouts by integrating existent design expertise. The proposed approach can automatically analyze legacy design data including circuits, layouts, and constraints, extract matched sub-circuits between new and legacy designs, and generate multiple layouts for the new design by utilizing the quality-approved legacy layouts as much as possible. Experimental results show that the proposed methodology can achieve high layout reusage rate, and hence the designers\' layout preference can be successfully reserved.
An EDA Framework for Large Scale Hybrid Neuromorphic Computing Systems
(2015)
Wen, Wei
;
Wu, Chi-Ruo
;
Hu, Xiaofang
;
Liu, Beiye
;
Ho, Tsung-Yi
In implementations of neuromorphic computing systems (NCS), memristor and its crossbar topology have been widely used to realize fully connected neural networks. However, many neural networks utilized in real applications often have a sparse connectivity, which is hard to be efficiently mapped to a crossbar structure. Moreover, the scale of the neural networks is normally much larger than that can be offered by the latest integration technology of memristor crossbars. In this work, we propose AutoNCS - an EDA framework that can automate the NCS designs that combine memristor crossbars and discrete synapse modules. The connections of the neural networks are clustered to improve the utilization of the memristor elements in crossbar structures by taking into account the physical design cost of the NCS. Our results show that AutoNCS can substantially enhance the utilization efficiency of memristor crossbars while reducing the wirelength, area and delay of the physical designs of the NCS.
PEGylation Site-Dependent Structural Heterogeneity Study of MonoPEGylated Human Parathyroid Hormone Fragment hPTH(1-34)
(2014-09-30)
Liu, Chih-Ying
;
Li, Xin
;
Chen, Wen-Yih
;
Chang, Li-Chiao
;
Chen, Yi-Fan
The structures of C- and N-terminally monoPEGylated human parathyroid hormone fragment hPTH(1-34) as well as their unmodified counterparts, poly(ethylene glycol) (PEG) and hPTH(1-34), have been studied by small-angle neutron scattering (SANS). The scattering results show that free hPTH(1-34) in 100 mM phosphate buffer (pH 7.4) aggregates into clusters. After conjugation with PEG, the PEG-peptide conjugates self-assemble into a supramolecular core-shell structure with a cylindrical shape. The PEG chains form a shell around the hPTH(1-34) core to shield hPTH(1-34) from the solvent. The detailed structural information on the self-assembled structures is extracted from SANS using a model of the cylindrical core with a shell of Gaussian chains attached to the core surface. On the basis of the data, because of the charge-dipole interactions between the conjugated PEG chain and the peptide, the conjugated PEG chain forms a more collapsed conformation compared to free PEG. Moreover, the size of the self-assembled structures formed by the C-terminally monoPEGylated hPTH(1-34) is about 3 times larger than that of the N-terminally monoPEGylated hPTH(1-34). The different aggregation numbers of the self-assembled structures, triggered by different PEGylation sites, are reported. These size discrepancies because of different PEGylation sites could potentially affect the pharmacokinetics of the hPTH(1-34) drug.