The modeling of p–InxGa1−xN/n–Si hetero junction diodes without using the buffer layer were investigated with the “top-top” electrode. The p–Mg-GaN and p–Mg-In0.05Ga0.95N were deposited directly on the n–Si (100) wafer by the RF reactive sputtering at 400 °C with single cermet targets. Al and Pt with the square size of 1 mm2 were used for electrodes of p–InxGa1−xN/n–Si diodes. Both devices had been designed to prove the p-type performance of 10% Mg-doped in GaN and InGaN films. By Hall measurement at the room temperature (RT), the holes concentration and mobility were determined to be Np = 3.45 × 1016 cm−3 and µ = 145 cm2/V·s for p–GaN film, Np = 2.53 × 1017 cm−3, and µ = 45 cm2/V·s for p–InGaN film. By the I–V measurement at RT, the leakage currents at −5 V and turn-on voltages were found to be 9.31 × 10−7 A and 2.4 V for p–GaN/n–Si and 3.38 × 10−6 A and 1.5 V for p–InGaN/n–Si diode. The current densities at the forward bias of 20 V were 0.421 and 0.814 A·cm−2 for p–GaN/n–Si and p–InGaN/n–Si devices. The electrical properties were measured at the temperature range of 25 to 150 °C. By calculating based on the TE mode, Cheungs’ and Norde methods, and other parameters of diodes were also determined and compared.

The all-sputtered Al/SiO2/p-GaN metal-oxide-semiconductor (MOS) Schottky diode was fabricated by the cost-effective radio-frequency sputtering technique with a cermet target at 400 °C. Using scanning electron microscope (SEM), the thicknesses of the electrodes, insulator SiO2 layer, and p-GaN were found to be ~250 nm, 70 nm, and 1 µm, respectively. By Hall measurement of a p-Mg-GaN film on an SiO2/Si (100) substrate at room temperature, the hole’s concentration (Np) and carrier mobility (μ) were found to be Np = 4.32 × 1016 cm−3 and μ = 7.52 cm2·V−1·s−1, respectively. The atomic force microscope (AFM) results showed that the surface topography of the p-GaN film had smoother, smaller grains with a root-mean-square (rms) roughness of 3.27 nm. By I–V measurements at room temperature (RT), the electrical properties of the diode had a leakage current of ~4.49 × 10−8 A at −1 V, a breakdown voltage of −6 V, a turn-on voltage of ~2.1 V, and a Schottky barrier height (SBH) of 0.67 eV. By C–V measurement at RT, with a frequency range of 100–1000 KHz, the concentration of the diode’s hole increased from 3.92 × 1016 cm−3 at 100 kHz to 5.36 × 1016 cm−3 at 1 MHz, while the Fermi level decreased slightly from 0.109 to 0.099 eV. The SBH of the diode at RT in the C–V test was higher than in the I–V test because of the induced charges by dielectric layer. In addition, the ideality factor (n) and series resistance (Rs) determined by Cheung’s and Norde’s methods, other parameters for MOS diodes were also calculated by C–V measurement at different frequencies.

We study streaming data for a data warehouse, which combines different sources. We consider the relative answers to OLAP queries on a schema, as distributions with the L1 distance and approximate the answers without storing the entire data warehouse. We first study how to sample each source and combine the samples to approximate any OLAP query. We then consider a streaming context, where a data warehouse is built by streams of different sources. We first show a lower bound on the size of the memory necessary to approximate queries and then consider a statistical hypothesis where some attributes determine fixed distributions of the measure. We use the sampling methods to learn the statistical model and approximate OLAP queries. In this case, we approximate OLAP queries with a finite memory. We apply the method to a dataset which simulates the data of sensors, which provide weather parameters over time and locations from different sources.