Revisiting the cutting of the firing squad synchronization

Various synchronization algorithms have been introduced in literature during the last decades to deal with the firing squad synchronization problem on cellular automata (CA). Among others defective CA algorithms, where the CA cell is able to transmit information without previous processing, have been also presented. In our case, originating from the classical Mazoyer’s paper, where a minimum-time solution is presented with 6 states, a one-dimensional CA where one cell may permanently fail is presented. In the proposed algorithm, the defective cell can neither process nor transmit information any longer, while it is considered that such dynamic defects may become apparent in any time step of computation. A thorough analysis of the synchronization, in terms of location and time at which cell fails, for the cells found in both sides of defective cell is delivered to decipher the corresponding maximal possible number of synchronized cells in each part of the cut, due to defect, CA array. The proposed algorithm is properly extended to consider more than one defective cells that may occur in the under study one-dimensional CA. Based on the aforementioned analysis, we provide the generalization of synchronization with multiple totally defective cells, while application examples of the generalized CA algorithm in case of two defective cells are also presented. Finally, another intriguing aspect refers to handling of states that could be tentatively characterized as unknown, in a confrontation similar to the previous defective state but also different, since now this(these) cell(s) are not stated as faulty but unknown. As a result, a new one-dimensional CA with less states, compared to the previous CA defective algorithms, able to synchronize the maximal possible number of cells in each part occurs.     

μARTMAP: use of mutual information for category reduction inFuzzy ARTMAP

A new architecture called muARTMAP is proposed to impact a category proliferation problem present in Fuzzy ARTMAP. Under a probabilistic setting, it seeks a partition of the input space that optimizes the mutual information with the output space, but allowing some training error, thus avoiding overfitting. It implements an inter-ART reset mechanism that permits handling exceptions correctly, thus using few categories, especially in high dimensionality problems. It compares favorably to Fuzzy ARTMAP and Boosted ARTMAP in several synthetic benchmarks, being more robust to noise than Fuzzy ARTMAP and degrading less as dimensionality increases. Evaluated on a real-world task, the recognition of handwritten characters, it performs comparably to Fuzzy ARTMAP, while generating a much more compact rule set.     

Stress-induced local trap levels in Au/n-GaAs Schottky diodes with embedded InAs quantum dots

Local trap levels in Au/n-GaAs Schottky diodes with embedded InAs quantum dots, generated after a long time of the device operation, have been investigated with low-frequency noise measurements performed in the temperature range of 77-298 K and at the forward current of 30 nA. Whereas the initial devices show a pure 1/f noise behavior, after a long time of operation, recombination noise was observed at frequencies above 100 Hz, in addition to the 1/f noise at lower frequencies. Analysis of the recombination noise data obtained on structures where different GaAs cap layer thicknesses have been removed by etching allowed us to determine the activation energy of the local traps and have a rough estimation of their spatial distribution.     

Islet amyloid polypeptide decreases the effects of insulin-like growth factor-I on glucose transport and glycogen synthesis in skeletal muscle

Previous studies have shown that islet amyloid polypeptide (IAPP) is co-secreted with insulin from the beta-cell. IAPP reduces insulin-stimulated rates of glycogen synthesis in skeletal muscle but the mechanisms are unclear. Insulin-like growth factor I (IGF-I) is an important regulator of glucose metabolism in skeletal muscle and acts through its own receptor, which has many structural and functional similarities with the insulin receptor. Despite this, the effects of IGF-I on glucose utilization are not identical to those of insulin. The aim of the study was to determine the effects of IAPP on IGF-I-stimulated rates of glucose transport and metabolism (measured by 3-O-methyl[3H]glucose and [U-14C]glucose, respectively) in rat soleus muscle, and compare them with those simulated by insulin. IAPP (10 nM) decreased the sensitivity of 3-O-methylglucose transport, the flux of glucose to hexosemonophosphate and the sensitivity of glycogen synthesis to IGF-I. In contrast, IAPP had no effect on IGF-I-stimulated rates of lactate formation (i.e., glycolysis). IAPP decreased the sensitivity of 3-O-methylglucose transport and glycogen synthesis to insulin. It is concluded that IAPP blunts the stimulation of glucose uptake and deposition by IGF-I or insulin in skeletal muscle. These observations expand those made initially for IAPP and insulin and suggest that IAPP affects IGF-I- or insulin-stimulated glucose metabolism in muscle by a mechanism which is common for both hormones. These experiments may serve as a framework for future studies in order to clarify the mechanisms by which IAPP affects glucose metabolism in skeletal muscle.     

Semi-Analytical Modeling of Short-Channel Effects in Si and Ge Symmetrical Double-Gate MOSFETs

A simple analytical expression of the 2-D potential distribution along the channel of silicon symmetrical double-gate (DG) MOSFETs in weak inversion is derived. The analytical solution of the potential distribution is compared with the numerical solution of the 2-D Poisson's equation in terms of the channel length L, the silicon thickness t _Si, and the gate oxide thickness t _OX. The obtained results show that the analytical solution describes, with good accuracy, the potential distribution along the channel at different positions from the gate interfaces for well-designed devices when the ratio of L/t _Si is ges 2-3. Based on the 2-D extra potential induced in the silicon film due to short-channel effects (SCEs), a semi-analytical expression for the subthreshold drain current of short-channel devices is derived. From the obtained subthreshold characteristics, the extracted device parameters of the subthreshold slope, drain-induced barrier lowering, and threshold voltage are discussed. Application of the proposed model to devices with silicon replaced by germanium demonstrates that the germanium DG MOSFETs are more prone to SCEs.     

Stability of Amorphous-Silicon and Nanocrystalline Silicon Thin-Film Transistors Under DC and AC Stress

Bottom-gated n-channel thin-film transistors (TFTs) were fabricated using hydrogenated amorphous-silicon (a-Si:H)/ nanocrystalline silicon (nc-Si:H) bilayers as channel materials, which are deposited by plasma-enhanced chemical vapor deposition at low temperatures. The stability of these devices is investigated under static and dynamic bias stress conditions. For comparison, the stability of a-Si:H and nc-Si:H single-layer TFTs is investigated under similar bias stress conditions. The overall results demonstrate that the a-Si:H/nc-Si:H bilayer TFTs are superior compared with their counterparts of a-Si:H and nc-Si:H TFTs regarding device performance and stability.     

Hysteresis effect in bottom-gate polymorphous silicon thin-film transistors

The hysteresis effect observed in the transfer characteristics of n-channel bottom-gate hydrogenated polymorphous silicon (pm-Si:H) thin-film transistors (TFTs) is investigated in terms of the channel width. Such phenomenon is observed in devices of wide channel (>20 μm), whereas it diminishes in devices of narrow channel. The hysteresis of wide channel TFTs is mainly due to charges injected from the channel, trapped in the gate dielectric. As the channel width is reduced the edge effect becomes more significant and the effect of carrier injection from the channel is eliminated, which is balanced by the effect of charge injection from the gate electrode.     

On-state drain current modeling of large-grain poly-Si TFTs based on carrier transport through latitudinal and longitudinal grain boundaries

An analytical on-state drain current model of large-grain polycrystalline silicon thin-film transistors (polysilicon TFTs) is presented, based on the carrier transport through latitudinal and longitudinal grain boundaries. The model considers an array of square grains in the channel, with the current flowing along the longitudinal grain boundaries or through the grains and across the latitudinal grain boundaries. Application of the proposed model to excimer lased annealed polysilicon TFTs reveals that, at low gate voltages in the moderate inversion region, the longitudinal grain boundaries influence the effective carrier mobility and the drain current. As the gate voltage increases, the latitudinal grain boundaries have larger impact to the current flow due to reduction of the potential barrier at the grain boundaries. The effect of the laser energy density on the quality of the grains and grain boundaries is investigated.     

An analytical hot-carrier induced degradation model in polysilicon TFTs

Hot-carrier effects in n-channel polysilicon thin-film transistors (TFTs), with channel width W=10 /spl mu/m and length L=10 /spl mu/m, are investigated. An analytical model predicting the post-stress performance is presented, by treating the channel of the stressed device as a series combination of a damaged region extended over a length /spl Delta/L beside the drain and a region of length L-/spl Delta/L having the properties of the unstressed device. The apparent channel mobility is derived considering that the mobility of the damaged region is described with the mobility of amorphous Si TFTs, whereas the mobility of the undamaged region is described with the mobility of the virgin device. From the evolution of the static characteristics during stress, the properties of the damaged region with stress time are investigated.     

Analytical modelling for the current-voltage characteristics of undoped or lightly-doped symmetric double-gate MOSFETs

An analytical drain current model for undoped (or lightly-doped) symmetric double-gate (DG) MOSFETs is presented. This model is based on the subthreshold leakage current in weak inversion due to diffusion of carriers from source to drain and an analytical expression for the drain current in strong inversion of long-channel DG MOSFETs, both including the short-channel effects. In the saturation region, the series resistance, the channel length modulation, the surface-roughness scattering and the saturation velocity effects were also considered. The proposed model has been validated by comparing the transfer and output characteristics with simulation and experimental results.