This paper reviews recent progress in silicon photonics and compares it with other optical device platforms. The key components for optical communication systems, including arrayed waveguide gratings, optical switches, modulators and optical functional devices fabricated on silicon photonics platforms are explained. The integration of III-V compounds, lithium niobate, polymers, phase change and other functional materials are necessary to strengthen silicon photonics platforms. These are also reviewed, and the feasibilities are discussed. Finally, I express my personal view as to the best research direction for silicon photonics.
A fourth-order flat passband bandpass filter (BPF) is realized using proposed resonators and half-wave open-looped resonators. The fourth-order flat passband BPF is necessary to prepare at least two kinds of resonators that have high and low unloaded quality factors. The proposed resonator is configured based on a half-wave hairpin-shaped resonator with a parallel-connected resistor. It is assigned as a resonator with a low unloaded quality factor. The advantage of the proposed resonator is flexible adjustment of the unloaded quality factor by changing the value and setting position of the resistor. Using equations and electromagnetic simulation, we show that the unloaded quality factor can be controlled by setting the position of the resistor. A half-wave open-looped resonator is assigned as a resonator with a high unloaded quality factor. Furthermore, a fourth-order flat passband BPF with transmission zeros and a flat passband is constructed on a printed circuit board by using the proposed resonators and half-wave open-looped resonators. Flat passband characteristics of a fourth-order BPF is satisfied with the design specifications.
The nonlinear behaviors of time-interleaved analog-to-digital converter (TIADC) caused by non-ideal circuit implementations degrade performance of TIADC significantly. This paper proposes a calibration method based on model inversion strategy to compensate for static nonlinearities in TIADC, since static nonlinearities are the major compositions of the nonlinear behaviors in most situations. The proposed periodic time-varying adaptive calibration method can estimate coefficients of inverse polynomial model directly and capture variations of the nonlinear parameters when environment changes. The method proposed in this paper is also applicable to dynamic nonlinearities by changing polynomial with Volterra series. Simulation results show the effectiveness and robustness of the proposed method.
The unwanted electro-forming process is unavoidable for the practical application of most resistive random access memory (ReRAM) devices, which is always being one of the obstacles for the massive commercialization of this novel electronic device. In this letter, a forming-free Pt/Ti/(TiO2-x)/Ta2O5/Pt based ReRAM device is demonstrated with an additional feature of the low operation voltage. The fitting result of the measured I-V curves reveals that resistive switching of the fabricated device is conducted by the electrons trapping/de-trapping process in deep-level electron traps of the TiO2-x layer, which is formed through the spontaneous “oxygen grabbing” reaction in the interface of Ti and Ta2O5 during the film deposition. The plentiful oxygen vacancy defects and the thin resistive switching zone (TiO2-x) ensure the forming-free and low operating voltage characteristics. Using “oxygen grabbing” to pre-produce abundant electron trapping centers for the resistive switching provides a simple way for the fabrication of the forming-free ReRAM device with low operating voltage, aiming to the high-density and low-power memory applications.
A linear electronically tunable sinusoid oscillator (LETSO) using a Current Feedback Amplifier (CFA) and three grounded capacitors, along with a pair of matched analog multiplier devices as the composite active building block (ABB) is presented. The design yields a high-Q filter response; thereby a sustained linear oscillatory response is implemented using the concept of Short-Circuit Natural Frequency (SCNF) at Q ∼∞. The oscillation frequency (fo) is linearly tunable by the multiplier control voltage (V). Effects of the CFA port-rolloff parameters and parasitic capacitors are negligible. The LETSO - response upto a range of fo∼ 12MHz with measured THD ∼ 2% and linearity error (Δ∼ 3.6%) had been experimentally verified.
In this letter, a machine learning (ML) model is presented to predict the variation of the threshold voltage (Vth) according to the taper angle and target word line (WLT) position in 3D NAND flash memory. Through Technology Computer-Aided Design (TCAD) simulation, Vth is extracted according to taper angle and WLT position. TCAD data is used as the training data set required for learning by an artificial neural network algorithm (NNA). The completed ML model is then used to predict Vth for each word line (WL). It was also confirmed that the ML model predicted well even for TCAD data that was not used as a training data set.