Nanoscale Research Letters · 2005 · 14 citations · 7 references
ComputingComputer ScienceFeature EngineeringMachine LearningTime Series AnalysisDigital Signal ProcessingPattern RecognitionRadar Signal ProcessingRadar SystemsSignal ProcessingTime-frequency AnalysisFrequency-domain ClassificationPulse CompressionRadar Classification Training DataSignal ClassificationSignal-to-noise RatioSignal-to-noise Ratio AnalysisTime-frequency Feature ExtractionTime-frequency Pattern Recognition
In this paper, pMOSFETs featuring atomic layer deposition (ALD) tungsten (W) using SiH<sub>4</sub> and B<sub>2</sub>H<sub>6</sub> precursors in 22 nm node CMOS technology were investigated. It is found that, in terms of threshold voltage, driving capability, carrier mobility, and the control of short-channel effects, the performance of devices featuring ALD W using SiH<sub>4</sub> is superior to that of devices featuring ALD W using B<sub>2</sub>H<sub>6</sub>. This disparity in device performance results from different metal gate-induced strain from ALD W using SiH<sub>4</sub> and B<sub>2</sub>H<sub>6</sub> precursors, i.e. tensile stresses for SiH<sub>4</sub> (~2.4 GPa) and for B<sub>2</sub>H<sub>6</sub> (~0.9 GPa).
7
A scaled conjugate gradient algorithm for fast supervised learning
Martin F. Møller · Neural Networks · 1993 · 4.1K citations