Showing papers by "Jong-Ho Lee published in 2018"

Scaling Effect on Silicon Nitride Memristor with Highly Doped Si Substrate.

Abstract: A feasible approach is reported to reduce the switching current and increase the nonlinearity in a complementary metal-oxide-semiconductor (CMOS)-compatible Ti/SiNx /p+ -Si memristor by simply reducing the cell size down to sub-100 nm. Even though the switching voltages gradually increase with decreasing device size, the reset current is reduced because of the reduced current overshoot effect. The scaled devices (sub-100 nm) exhibit gradual reset switching driven by the electric field, whereas that of the large devices (≥1 µm) is driven by Joule heating. For the scaled cell (60 nm), the current levels are tunable by adjusting the reset stop voltage for multilevel cells. It is revealed that the nonlinearity in the low-resistance state is attributed to Fowler-Nordheim tunneling dominating in the high-voltage regime (≥1 V) for the scaled cells. The experimental findings demonstrate that the scaled metal-nitride-silicon memristor device paves the way to realize CMOS-compatible high-density crosspoint array applications.

Spiking Neural Network Using Synaptic Transistors and Neuron Circuits for Pattern Recognition With Noisy Images

Abstract: We demonstrate the hardware implementation of spiking neural network (SNN) with synaptic transistors and neuron circuits. The method of conversion from software fully-connected network (FCN) to hardware SNN with little degradation is discussed. The degradation of classification accuracy is analyzed in terms of device variation and noisy images. In addition, the accuracy degradation is significantly improved by stacking denoising autoencoder (DAE) layer. FCN–SNN conversion with very little performance drop is demonstrated using weight normalization, and SNN with DAE layer shows a great tolerance to input image noise.

Demonstration of Unsupervised Learning With Spike-Timing-Dependent Plasticity Using a TFT-Type NOR Flash Memory Array

Abstract: We investigate the characteristics of a synaptic imitation device using a thin-film transistor (TFT)-type NOR flash memory cell with a half-covered floating gate. The long-term potentiation (LTP) and long-term depression (LTD) required for the operation of the spike-timing-dependent plasticity (STDP) algorithm are implemented using the proposed pulse scheme. Unsupervised learning is successfully demonstrated by applying the STDP learning rule through software MATLAB simulation reflecting the LTP/LTD characteristics of the fabricated TFT-type NOR flash memory array. We present the learning and recognition processes of $28\times28$ MNIST handwritten digit patterns. First, STDP learning in a single-neuron string ( $784\times1$ ) is investigated, after which STDP learning is demonstrated in a multineuron array ( $784\times10$ ) with a lateral inhibition function to demonstrate the ability of multipattern learning and recognition. Meanwhile, we investigate the key factors of STDP unsupervised learning. Finally, an approach is suggested to implement a hardware neural network using the conventional CMOS technology for STDP unsupervised learning as a visual pattern recognition system.

Observation of physisorption in a high-performance FET-type oxygen gas sensor operating at room temperature

Abstract: Oxygen (O2) sensors are needed for monitoring environment and human health. O2 sensing at low temperature is required, but studies are lacking. Here we report, for the first time, that the performance of a field effect transistor (FET)-type O2 sensor operating at 25 °C was improved greatly by a physisorption sensing mechanism. The sensing material was platinum-doped indium oxide (Pt–In2O3) nanoparticles formed by an inkjet printer. The FET-type sensor showed excellent repeatability under a physisorption mechanism and showed much better sensing performance than a resistor-type sensor fabricated on the same wafer at 25 °C. The sensitivity of the sensor increased with increasing Pt concentration up to ∼10% and decreased with further increasing Pt concentration. When the sensing temperature reached 140 °C, the sensing mechanism of the sensor changed from physisorption to chemisorption. Interestingly, the pulse pre-bias before the read bias affected chemisorption but had no effect on physisorption.

Aligned laminin core-polydioxanone/collagen shell fiber matrices effective for neuritogenesis

Abstract: Neural tissue regeneration is a significant challenge, because severe nerve injury is quite difficult to regenerate spontaneously. Although, many studies have been devoted to promote nerve regeneration, there are still many technical challenges to achieve satisfactory results. In this study, we designed biomimetic matrices composed of aligned laminin core-polydioxanone/collagen shell (Lam-PDO/Col) fibers, which can provide both topographical and biochemical cues for promoting neuritogenesis. The aligned Lam-PDO/Col core-shell fiber matrices were fabricated by magnetic field-assisted electrospinning with the coaxial system, and their potential as biofunctional scaffolds for promoting neuritogenesis was explored. It was demonstrated that the aligned Lam-PDO/Col core-shell fibers were successfully fabricated, and the laminin in the core of fibers was steadily and continuously released from fibers. In addition, the cellular behaviors of hippocampal neuronal cells on the matrices were significantly enhanced. Moreover, the aligned Lam-PDO/Col fiber matrices effectively improved and guided neurite outgrowth as well as the neurogenic differentiation by providing both topographical and biochemical cues through aligned fiber structure and sustained release of laminin. Collectively, it is suggested that the aligned Lam-PDO/Col core-shell fiber matrices are one of the most promising approaches for promoting neuritogenesis and neural tissue regeneration.

The effects of Jerusalem artichoke and fermented soybean powder mixture supplementation on blood glucose and oxidative stress in subjects with prediabetes or newly diagnosed type 2 diabetes

Abstract: The objective of this study was to evaluate the effect of supplementation with a Jerusalem artichoke and fermented soybean powder mixture on blood glucose and oxidative stress levels. This randomized, double-blinded, placebo-controlled study was conducted on 60 subjects with impaired fasting glucose (IFG), impaired glucose tolerance (IGT), or newly diagnosed type 2 diabetes. The subjects were randomly assigned to either a group that ingested 40 g of a Jerusalem artichoke and fermented soybean powder mixture (19.45 g each) daily or a group that received a placebo for 12 weeks. Paired t-test and independent t-test were performed for comparisons within groups and between groups, respectively. Supplementation with the Jerusalem artichoke and fermented soybean powder mixture reduced the levels of fasting glucose (p < 0.001) and FFAs (p = 0.034), glucose at 60 min (p = 0.004), glucose (p = 0.006) areas under the response curve (AUC), homeostasis model assessment-insulin resistance (p = 0.018), and the urinary 8-epi-prostaglandin F2α (8-epi-PGF2α) level (p = 0.028). The changes (Δ) in urinary 8-epi-PGF2α, glucose at 60 min, 120 min, and AUC, FFAs at 0 min and AUC were significantly different between the two groups. In addition, Δ glucose at 120 min (r = 0.472, p = 0.027) and the Δ glucose AUC (r = 0.572, p = 0.005) were positively correlated with △ plasma malondialdehyde in the test group. The consumption of a Jerusalem artichoke and fermented soybean powder mixture for 12 weeks was effective for reducing postprandial glucose and oxidative stress level, particularly 8-epi-PGF2α, in subjects with IFG, IGT, or newly diagnosed type 2 diabetes.

Neuromorphic Technology Based on Charge Storage Memory Devices

Abstract: Four synaptic devices are introduced for spiking neural networks (SNNs) and deep neural networks (DNNs). Unsupervised learning is successfully demonstrated by applying the STDP learning rule reflecting the LTP/LTD characteristics of the fabricated TFT-type NOR flash memory cells. Gated Schottky diode (GSD) and vertical NAND flash cell are proposed as synaptic device for DNNs. Using matched simulation, we obtained higher learning accuracy with GSD and NAND synaptic devices compared to that with a memristor-based synapse. Measured synaptic properties of the vertical NAND cells are reported for the first time.

A 16Gb 18Gb/S/pin GDDR6 DRAM with per-bit trainable single-ended DFE and PLL-less clocking

Abstract: Starting at 512Mb 6Gb/s/pin [1], GDDR5's speed and density have been steadily developing for about 10 years; recently achieving 8Gb 9Gb/s/pin [2] with per-pin timing training. Although 8Gb GDDR5X can operate at 12Gb/s [3] by increasing the burst length (BL) from 8 to 16, a degradation in system performance at a data granularity of 64B is seen. The I/O specification, using PLL clocking that additionally causes PLL jitter, has not changed much compared with GDDR5. To overcome these issues, GDDR6 introduced a dual channel for a data granularity of 32B with a BL16, per-bit training of l/ REF , and an equalizer with PLL-less clocking. This paper presents a 16Gb 18Gb/s/pin GDDR6 DRAM with a die architecture and high-speed circuit techniques on 1.35V DRAM process.

A Split-Gate Positive Feedback Device With an Integrate-and-Fire Capability for a High-Density Low-Power Neuron Circuit.

Abstract: Hardware-based spiking neural networks (SNNs) to mimic biological neurons have been reported. However, conventional neuron circuits in SNNs have a large area and high power consumption. In this work, a split-gate floating-body positive feedback (PF) device with a charge trapping capability is proposed as a new neuron device that imitates the integrate-and-fire function. Because of the PF characteristic, the subthreshold swing (SS) of the device is less than 0.04 mV/dec. The super-steep SS of the device leads to a low energy consumption of ~0.25 pJ/spike for a neuron circuit (PF neuron) with the PF device, which is ~100 times smaller than that of a conventional neuron circuit. The charge storage properties of the device mimic the integrate function of biological neurons without a large membrane capacitor, reducing the PF neuron area by about 17 times compared to that of a conventional neuron. We demonstrate the successful operation of a dense multiple PF neuron system with reset and lateral inhibition using a common self-controller in a neuron layer through simulation. With the multiple PF neuron system and the synapse array, on-line unsupervised pattern learning and recognition are successfully performed to demonstrate the feasibility of our PF device in a neural network.

An FET-type gas sensor with a sodium ion conducting solid electrolyte for CO2 detection

Abstract: In this paper, a Field-Effect-Transistor (FET)-type gas sensor using the composite of Na2CO3 and NaNO2 as the sensing material is investigated. The sensor has a floating gate (FG) passivated by an insulator stack and a control gate (CG), which are formed in an interdigitated form in a horizontal direction. The inkjet printing process forms the sensing material on the interdigitated FG and CG. Just before forming the sensing material, a (3-aminopropyl) triethoxysilane (APTES) monolayer is formed to prevent diffusion of Na+ ions contained in the sensing material through the insulator layer. The sensor can be electrically programmed like a typical memory device due to the FG, which makes the calibration of the sensor possible. CO2 gas sensing properties of the sensor are characterized at an operating temperature of 160 °C. The results indicate that the proposed sensor produces a reasonable and repeatable response to a certain concentration of CO2.

Graphene electrode with tunable charge transport in thin-film transistors

Abstract: Graphene, a single atomic layer of sp2-hybridized carbon, has immense potential as a transparent conducting material in electronic applications owing to its superior properties, including optical transparency and high conductivity. Particularly, the tunable work function of graphene enables the integration of graphene electrodes with various electronic devices. To achieve high performance in graphene-based devices, effective charge transport between the graphene electrode and the semiconducting material needs to be optimized; this is closely related to the modulation of the Schottky barrier (SB). In this study, we investigate the tunable charge transport properties as a function of graphene doping in n-channel thin-film transistors (TFTs) in terms of the electrical characteristics and low-frequency noise (LFN) behaviors. Alkali metal carbonates tuned the work function of graphene, resulting in a dramatic decrease in the SB and an improvement of the carrier injection in n-channel TFTs. The electrical performance of the TFTs was evaluated by extraction of the field-effect mobilities and ratio of contact resistance to total resistance. Furthermore, the level of contact noise created by the barrier height fluctuation and relative contribution of channel noise and contact noise in the TFTs was investigated by LFN measurements to demonstrate the tunable charge transport. Our findings therefore provide new insights into the tunable charge transport mechanism in graphene-based devices and reveal the immense potential of graphene as electrodes in high performance flexible and transparent displays.

An accurate and stable humidity sensing characteristic of Si FET-type humidity sensor with MoS 2 as a sensing layer by pulse measurement

Abstract: In this work, we demonstrate the pulse scheme to obtain stable humidity sensing characteristics in Si FET-type humidity sensor using MoS2 film as a sensing layer. To investigate the reaction between H2O molecules and MoS2 film, the transfer characteristics (ID-VCG) and transient drain current behaviors (ID-t) are measured in both pMOSFET and nMOSFET sensors by DC measurement. To verify the effect of the pulse scheme, the pulsed I–V (PIV) and the ID-t are measured as a parameter of relative humidity. The ID drift of the FET-type sensor is effectively eliminated by applying the pulse to the control-gate (CG) of the FET-type humidity sensor.

Association of toothbrushing and proximal cleaning with periodontal health among Korean adults: Results from Korea National Health and Nutrition Examination Survey in year 2010 and 2012.

Abstract: Aim Toothbrushing (TB), dental flossing (DF) and inter-dental brushing (IDB) are regarded as fundamental self-care methods for periodontal health. Few evidences on its effectiveness on periodontal health are available. Hence, this study aimed to evaluate the association of TB, DF, IDB and interaction effect with periodontal health. Materials and methods The nationally representative 4,766 Korean adults aged 19 years and older were cross-sectionally surveyed in 2010 and 2012. Periodontal health was defined as Community Periodontal Index 1-2 for gingivitis and 3-4 for periodontitis. The information about variables was from interview and blood analyses. Multivariable logistic regression analyses and the interaction effect between TB and proximal cleaning (PC: DF and/or IDB) were applied. Results Toothbrushing thrice or more per day and DF were associated with a lower prevalence of periodontitis by both 44%, while the preventive fraction of DF on gingivitis was 30%. The preventive fraction of interaction effects between TB thrice or more and PC were 78% for periodontitis and 68% for gingivitis among 40-59 year age group. Conclusions Toothbrushing and PC are independently associated with periodontal health. Hence, periodontists should recommend TB thrice or more per day and PC such as DF and IDB to promote periodontal health.

Space Program Scheme for 3-D NAND Flash Memory Specialized for the TLC Design

Abstract: A new space program (PGM) scheme is proposed to achieve reliable triple-level-cell (TLC) 3-D NAND flash memory. Considering the lateral diffusion issue of stored electrons in the nitride storage layer, the proposed scheme stores electrons in the nitride layer of the space region between adjacent cells to suppress the lateral movement of trapped electrons in the programmed target cells. The effect of the space PGM can be sustained until 104 s at 90 °C and up to 1k read cycles at 25 °C. The programmed space region of the nitride layer improves the retention characteristics of the cells in the PGM state by 40% and remarkably reduces the V th redistribution.

A Si FET-type Gas Sensor with Pulse-driven Localized Micro-heater for Low Power Consumption

Abstract: A poly-Si localized micro-heater for Si FET-type gas sensor is proposed. The gas sensor has an air gap under the heater to prevent the heat dissipation. It is verified that the heater temperature can be read by reading the heater resistance immediately after the heating pulse is turned on or off. A heater temperature of 112°C is achieved at a heat pulse bias of 2 V, which consumes ∼0.92 mW. The heating and cooling times of the heater are $\sim\ 200\mu\mathrm{s}$ and $\sim\ 100\mu\mathrm{s}$ , respectively. NO 2 and H 2 S sensing are successfully performed by using a pulse-driven micro-heater in the proposed gas sensor.

Hardware-based Neural Networks using a Gated Schottky Diode as a Synapse Device

Abstract: A gated Schottky diode is proposed for high-performance synapse devices and a means of designing a neural network using this device is described. The proposed gated Schottky diode operates in the saturation region with respect to the input voltage and is therefore immune to input noise and enables accurate vector-by-matrix multiplication. Moreover, by applying identical pulses to the bottom gate to store charges in a storage layer, the reverse saturation current increases almost linearly. Considering these special characteristics, we propose an architecture that uses a time-modulated input pulse and a learning rule based on a single conductance step. A three-layer perceptron network is trained using the conductance response of the synapse device and unidirectional weight-updating methods. In simulations using this network, the classification accuracy rate of MNIST training sets was found to be 94.50%. Compared to memristive devices, the improved linearity of the conductance response in our device is evidence of its higher accuracy.

Identification of human papillomavirus (HPV) subtype in oral cancer patients through microarray technology

Abstract: Human papilloma virus (HPV) is the main source of cervical cancer. Many recent studies have revealed the prevalence and prognosis of HPV associated with oropharyngeal squamous cell carcinoma, but fewer reports have evaluated HPV in oral squamous cell carcinoma (OSCC). The purpose of this study was to determine the prevalence and prognosis of HPV associated with OSCC according to HPV and tumor types. We used a DNA chip kit (MY-HPV chip kit ®, Mygene Co., Korea) to detect high-risk HPV subtypes (16, 18, 31, 33, 35, 39, 45, 51, 52, 54, 56, 58) and low-risk subtypes (6, 11, 34, 40, 42, 43, 44) among 187 patients. The prevalence was determined by Chi-square and Fisher’s exact tests, and the prognosis was calculated by the Kaplan–Meier method and the log-rank test. The overall prevalence of HPV in OSCC was 7.0% for all HPV positives and 4.3% for high-risk HPV positives. The prevalence of HPV was significantly higher in individuals under 65 years old and in those with tumors in the tongue and gum regions. The prognosis did not differ between the HPV-positive and -negative groups. Although the prevalence of HPV-positive cases in OSCC was low (7.0, 4.3%) and the prognosis did not depend on HPV positivity, HPV-associated OSCC should be considered in the evaluation and treatment of oral cancer patients. In addition, separating high- and low-risk groups based on the HPV status of other body parts might not be appropriate. The DNA microarray method can accurately detect known HPV subtypes simultaneously, but has limitations in detecting new subtypes. Vaccines can also be used to prevent HPV-associated OSCC in patients, so further studies on the prognosis and efficacy of vaccines should be undertaken.

Extracorporeal cardiopulmonary resuscitation in refractory intra-operative cardiac arrest: an observational study of 12-year outcomes in a single tertiary hospital.

Abstract: Refractory intra-operative cardiac arrest is a challenging issue for anaesthetists. In this study, we analysed the outcomes of adult patients who received extracorporeal cardiopulmonary resuscitation for refractory intra-operative cardiac arrest between 2005 and 2016, using data from our institutional extracorporeal membrane oxygenation registry. We defined refractory intra-operative cardiac arrest as the failure of a return of spontaneous circulation after 30 min of cardiopulmonary resuscitation. The primary outcome measure was neurologically intact survival with a cerebral performance category score of 1 or 2 at hospital discharge. Between 2005 and 2016, extracorporeal cardiopulmonary resuscitation was used to treat 23 patients who experienced refractory cardiac arrest in the operating room. The survival rates of neurologically-intact subjects were 9/23 (39%) and 6/23 (26%) at 24 h postoperatively and at hospital discharge, respectively. The main cause of refractory-intra-operative cardiac arrest was haemorrhagic shock in 13 out of 23 (57%) patients, and the neurologically-intact survival rate in these patients was 3/13 (23%) at discharge. Our study showed that approximately a quarter of patients with refractory intra-operative cardiac arrest caused by haemorrhage would receive survival benefit from extracorporeal cardiopulmonary resuscitation. Therefore, extracorporeal cardiopulmonary resuscitation may be a possible option in this clinically-challenging situation.

Measurement of hepatitis B virus DNA in fresh versus processed dentin from chronically infected patients

Abstract: Demineralized dentin matrix (DDM) is commonly used as a bone-graft substitute. This study measured and compared human hepatitis B viruses (HBV) DNA in fresh dentin to that of dentin processed into DDM extracted during dental treatment from HBV-infected patients. The hypothesis was that the processing procedure for DDM would inactivate or eliminate HBV in the dentin matrix obtained from infected patients. Dentin from eighteen HBV-infected patients was collected and each dentin specimen was divided into two fragments. One fragment was used before processing as fresh dentin (control group) and the other was processed into DDM (experimental group). DNA was extracted and purified from each fresh and processed dentin specimen and the HBV DNA copy number quantitated by real time polymerase chain reaction. The HBV DNA copy number in the fresh dentin specimens were compared relative to serologic test results. The second parameter was to evaluate the effectiveness of the processing procedure (defatting, demineralization, freeze-drying, and sterilization) to inactivate or eliminate HBV by comparing the DNA copy number in the processed DDM with that in the matched fresh dentin specimens. All results were analyzed using Mann–Whitney U test to compare numerical measurements between groups and differences were considered statistically significant at P-values less than 0.05. The presence of HBV DNA was detected in 55.56% (10/18) of the fresh dentin specimens. For the ten HBV DNA-positive fresh dentin specimens, HBV DNA was detected in two (20%) of the matched processed dentin specimens. The copy number of HBV DNA in the two positive processed dentin specimens was 1.79 and 4.03, which were statistically lower than that of the fresh dentin specimens (P = 0.0167). The results from this study suggested that fresh dentin may be a carrier of HBV and that the procedure used to generate DDM extensively reduced the levels of HBV DNA. Further studies are needed to evaluate the infectivity of HBV in processed dentin.

Novel Boosting Scheme Using Asymmetric Pass Voltage for Reducing Program Disturbance in 3-Dimensional NAND Flash Memory

Abstract: In this paper, novel boosting scheme using asymmetric pass voltage ( $\text{V}_{\mathrm{ pass}}$ ) is proposed to obtain high channel boosting potential and to reduce program disturbance in 3-D NAND flash memory. The proposed scheme has the same program bias and timing conditions as conventional self-boosting except for $\text{V}_{\mathrm{ pass}}$ voltages applied to both adjacent word-lines of selected word-line (WLsel). Reduced $\text{V}_{\mathrm{ pass}}$ ( $\text{V}_{\mathrm{ pass1}} =\,\,\text{V}_{\mathrm{ pass}} - {\Delta }\text{V}$ ) is applied to previous word-line (WL $_{\rm n-{1}}$ ) of WLsel and increased $\text{V}_{\mathrm{ pass}}$ ( $\text{V}_{\mathrm{ pass2}} =\,\,\text{V}_{\mathrm{ pass}}+{\Delta }\text{V}$ ) is applied to next word-line (WL $_{\rm n+{1}}$ ). In this scheme, the $\text{V}_{\mathrm{ pass1}}$ cuts the channel off and causes local boosting when the channel potentials of inhibit strings are boosted up. Meanwhile, the $\text{V}_{\mathrm{ pass2}}$ compensates the program speed reduction of selected cell (cellsel) induced by the decreased voltage of the $\text{V}_{\mathrm{ pass1}}$ . Through the measurements of program disturbance in fabricated devices, it is revealed that the program disturbance is significantly improved without the reduction of program speed by the proposed scheme. Furthermore, the $\text{V}_{\mathrm{ pass1}}$ and $\text{V}_{\mathrm{ pass2}}$ are optimized to maximize the improvement.

Per-oral cross-facial sural nerve graft for facial reanimation

Abstract: Cross-facial nerve graft is considered the treatment of choice for facial reanimation in patients with unilateral facial palsy caused by central facial nerve damage. In most cases, a traditional parotidectomy skin incision is used to locate the buccal and zygomatic branches of the facial nerve. In this study, cross-facial nerve graft with the sural nerve was planned for three patients with facial palsy through an intraoral approach. An incision was made on the buccal cheek mucosa, and the dissection was performed to locate the buccal branch of the facial nerve. The parotid papillae and parotid duct were used as anatomic landmarks to locate the buccal branch. The intraoral approach is more advantageous than the conventional extraoral approach because of clear anatomic marker (parotid papilla), invisible postoperative scar, reduced tissue damage from dissection, and reduced operating time.

Characterization of Primary Epithelial Cells Derived from Human Salivary Gland Contributing to in vivo Formation of Acini-like Structures.

Abstract: Patients with head and neck cancer are treated with therapeutic irradiation, which can result in irreversible salivary gland dysfunction. Because there is no complete cure for such patients, stem cell therapy is an emerging alternative for functional restoration of salivary glands. In this study, we investigated in vitro characteristics of primarily isolated epithelial cells from human salivary gland (Epi-SGs) and in vivo formation of acini-like structures by Epi-SGs. Primarily isolated Epi-SGs showed typical epithelial cell-like morphology and expressed E-cadherin but not N-cadherin. Epi-SGs expressed epithelial stem cell (EpiSC) and embryonic stem cell (ESC) markers. During long-term culture, the expression of EpiSC and ESC markers was highly detected and maintained within the core population with small size and low cytoplasmic complexity. The core population expressed cytokeratin 7 and cytokeratin 14, known as duct markers indicating that Epi-SGs might be originated from the duct. When Epi-SGs were transplanted in vivo with Matrigel, acini-like structures were readily formed at 4 days after transplantation and they were maintained at 7 days after transplantation. Taken together, our data suggested that Epi-SGs might contain stem cells which were positive for EpiSC and ESC markers, and Epi-SGs might contribute to the regeneration of acini-like structures in vivo. We expect that Epi-SGs will be useful source for the functional restoration of damaged salivary gland.

Memory device for controlling refreshing operation

Abstract: Provided is a memory device capable of reducing power consumption. The memory device includes a plurality of memory cells; and a self refresh controller configured to perform a refreshing cycle, which includes a first time interval and a second time interval, for a plurality of number of times, the second time interval being longer than the first section, wherein the self refresh controller is configured to perform a burst refreshing operation during the first time interval and to perform a power supply controlling operation during the second time interval.

A Novel Analysis of ${L}_{\text{gd}}$ Dependent-1/ ${f}$ Noise in In 0.08 Al 0.92 N/GaN

Abstract: In0.08Al0.92N/GaN fin-type high-electron mobility transistors (fin-HEMTs) with different gate-to-drain lengths ( ${L} _{\text {gd}}$ ) are fabricated and characterized by dc and low-frequency noise (LFN) measurements. The fabricated device with the largest ${L} _{\text {gd}}$ exhibits the degradation of the maximum drain current and transconductance with a positive shift of the threshold voltage. LFN measurements of the In0.08Al0.92N/GaN fin-HEMTs reveal clear 1/ ${f}$ behavior of the noise spectra, and the minimum value is observed in the device at ${L} _{\text {gd}} = {20}\,\,\mu \text{m}$ . The devices with smaller ${L} _{\text {gd}}$ follow a carrier number fluctuation noise model owing to electron trapping/detrapping into the In0.08Al0.92N barrier layer from the 2-D electron gas (2DEG) channel. In contrast, the device with the largest ${L} _{\text {gd}}$ shows correlated mobility fluctuations due to the large 2DEG mobility fluctuations in the large access area.

Erratum: Natural killer cell activity and interleukin-12 in metabolically healthy versus metabolically unhealthy overweight individuals (Front. Immunol. (2017) 8 (1700))

Abstract: [This corrects the article DOI: 10.3389/fimmu.2017.01700.].

A source drain symmetric and interchangeable bidirectional tunneling field effect transistor

Abstract: A novel bidirectional tunneling field effect transistor with source drain symmetric and interchangeable functions (SDSI BT FET) is proposed in this work. Different from the conventional tunneling field effect transistors (TFET), once the absolute value of drain-to-source bias voltage is determined, no matter which of the source drain interchangeable electrode is set to be drain, the proposed SDSI BT FET brings stable and unified transfer characteristics, thereafter, shows the source drain symmetric and interchangeable functions, which makes it more compatible with CMOS circuit compared to conventional TFET. It also has inherited the advantages of conventional TFET including lower subthreshold swing, lower static power dissipation, etc. Furthermore, through design optimization, a lower reverse biased leakage current and higher Ion-Ioff ratio can be observed.

Reconfigurable Cell String Having FET and Super-Steep Switching Diode Operation in 3D NAND Flash Memory

Abstract: A new reconfigurable cell string based on 3D NAND flash cell string is proposed and investigated by using TCAD simulation tool. By adapting ambipolar contact, the cell string has switchable characteristics between diode-type and FET-type. The diode-type operation has a steep subthreshold swing (SS) less than 1 mV/dec and the FET-type operation has comparable performances to conventional 3D NAND flash cell string.

Unsupervised Online Learning With Multiple Postsynaptic Neurons Based on Spike-Timing-Dependent Plasticity Using a TFT-Type NOR Flash Memory Array.

Abstract: We present a two-layer fully connected neuromorphic system based on a thin-film transistor (TFT)-type NOR flash memory array with multiple postsynaptic (POST) neurons. Unsupervised online learning by spike-timing-dependent plasticity (STDP) on the binary MNIST handwritten datasets is implemented, and its recognition result is determined by measuring firing rate of POST neurons. Using a proposed learning scheme, we investigate the impact of the number of POST neurons in terms of recognition rate. In this neuromorphic system, lateral inhibition function and homeostatic property are exploited for competitive learning of multiple POST neurons. The simulation results demonstrate unsupervised online learning of the full black-and-white MNIST handwritten digits by STDP, which indicates the performance of pattern recognition and classification without preprocessing of input patterns.