A Fast Algorithm for Wave Propagation from a Plane or a Cylindrical Surface

The newly developed Taylor-Interpolation-FFT (TI-FFT) algorithm dramatically increases the computational speeds for millimeter wave propagation from a planar (cylindrical) surface onto a “quasi-planar” (“quasi-cylindrical”) surface. Two different scenarios are considered in this article: the planar TI-FFT is for the computation of the wave propagation from a plane onto a “quasi-planar” surface and the cylindrical TI-FFT is for the computation of wave propagation from a cylindrical surface onto a “quasi-cylindrical” surface. Due to the use of the FFT, the TI-FFT algorithm has a computational complexity of O(N ²?log₂? N ²) for an N?×?N computational grid, instead of N ⁴ for the direct integration method. The TI-FFT algorithm has a low sampling rate according to the Nyquist sampling theorem. The algorithm has accuracy down to ?80 dB and it works particularly well for narrow-band fields and “quasi-planar” (“quasi-cylindrical”) surfaces.     

Study of the correlation properties of the surface structure of <Emphasis Type="Italic">nc</Emphasis>-Si/<Emphasis Type="Italic">a</Emphasis>-Si:H films with different fractions of the crystalline phase

The correlation properties of the structure of nc-Si/a-Si:H films with different volume fractions of the crystalline phase are studied using 2D detrended fluctuation analysis. Study of the surface relief of experimental samples showed that with increasing in volume fraction of the crystalline phase in the nc-Si/a-Si:H films, the size and number of nanoclusters on their surface grow. The size of Si nanocrystals in the a-Si:H matrix (6–8 nm) indicates the formation of coarse nanoclusters due to the self-organization of Si nanocrystals in groups under laser radiation. According to 2D detrended fluctuation analysis data, the number of correlation vectors (harmonic components) in the nc-Si/a-Si:H film structure increased with an increase in the nanocrystal fraction in the films.     

An improved proportionate least mean <Emphasis Type="Italic">p</Emphasis>-power algorithm for adaptive filtering

The least mean p-power error criterion has been successfully used in adaptive filtering due to its strong robustness against large outliers. In this paper, we develop a new adaptive filtering algorithm, named the proportionate least mean p-power (PLMP) algorithm, which uses the mean p-power error as the adaptation cost function. Compared with the standard proportionate normalized least mean square algorithm, the PLMP can achieve much better performance in terms of the mean square deviation, especially in the presence of impulsive non-Gaussian noises. The mean and mean square convergence of the proposed algorithm are analyzed, and some related theoretical results are also obtained. Simulation results are presented to verify the effectiveness of our proposed algorithm.     

Electronic and optical properties of AlAs/Al<Subscript>x</Subscript>Ga<Subscript>1−<Emphasis Type="Italic">x</Emphasis></Subscript>As(110) superlattices

Electronic states in the conduction band of (AlAs)_M(Al_xGa_1?xAs)_N(110) superlattices are investigated for various M and N. It is shown that electronic properties of these structures are mainly determined by electrons of two pairs of valleys, namely, either Γ-X ^Z or X ^X –X ^Y . Calculations based on the developed model of joining the envelope functions were carried out. Miniband spectra, symmetry, and localization of wave functions, as well as probabilities of miniband-to-miniband infrared absorption, are determined and analyzed. It is shown that, in the case of the X ^X –X ^Y pair of valleys, the absorption probabilities are high not only for polarization of light along the growth axis of the superlattice but also for the normal incidence of an optical wave on the structure surface.     

Frequency-time codes based on the decimation method

A regular rule for building uniform frequency-time codes, having length p and a large volume, over the Galois prime fields GF(p) based on the decimation method has been proposed. It was shown that the noise immunity coefficient of the decimation code proposed was higher than the uniform Reed-Solomon codes built over the corresponding fields GF(p).     

Performance modeling and analysis of hypoexponential network servers

Hypoexponential servers are commonly seen in today’s computer and communication networks whereby incoming packets are processed by the network server in multiple stages with each stage having a different processing time. This paper presents an analytical model to capture the behavior and subsequently analyze the performance of these network servers or similarly behaving systems. From our model, we derive key performance measures and features which include CPU utilization, system idleness, mean throughput, packet loss, mean system and queuing packet delays, and mean system and queue sizes. In addition, we present two popular finite queueing models (namely, M / D / 1 / K and M / M / 1 / K) to approximate our hypoexponential model. Results show that the both of these approximate models give close results when the system queue size is large.     

On the thermopower and thermomagnetic properties of Er<Subscript><Emphasis Type="Italic">x</Emphasis></Subscript>Sn<Subscript>1–<Emphasis Type="Italic">x</Emphasis></Subscript>Se solid solutions

The Er _x Sn_1–x Se system is characterized by a significant deviation of the temperature dependence of the differential thermopower from linearity at temperatures below room temperature and a change in the sign of the thermomagnetic coefficient. The deviation of the thermopower of Er _x Sn_1–x Se samples in the nonequilibrium state from linearity is found to be caused mainly by the entrainment of charge carriers by phonons α_ph. The statistical forces of electronic entrainment, A _ph(ε), are estimated.     

Influence of overhead on LTE downlink performance: a comprehensive model

Overhead resource elements (REs) in Long Term Evolution (LTE) networks are used for some control, signaling and synchronization tasks at both the Physical level and Media Access Control sub-level. Accurately computing all the overhead REs is necessary to achieve an efficient system design, which is difficult because LTE is a complex standard that contains a large number of implementation flexibilities and system configurations. The number of such REs depends on both the system configurations and services demanded. Aiming at exploring the influence of overhead on LTE downlink performance, we first parametrize each system configuration—including parameters corresponding to enhancement techniques such as Adaptive Modulation and Coding and Multi-Antenna Transmissions techniques—and those of the resource allocation mechanisms (which depend on users’ services). Second, using such parametrization, we model all overheads for synchronization, controlling and signaling operations in LTE Physical Downlink Shared/Control Channels. This allows for dynamically computing the useful REs (by subtracting the overhead REs from the total ones), both per Transmission Time Interval (TTI) and per frame (and hence, the corresponding bit rates). Our data rate-based performance model is able to accurately compute: (1) the real, exact system data rate or “throughput” (instead of approximations); and (2) the maximum number of simultaneous multi-service users per TTI that is able to support (called here “capacity”). Aiming at understanding the impact of each overhead mechanism, we have carried out a variety of simulations, including different service provision scenarios, such as multi-user with multi-application. The simulation results prove our starting hypothesis that the influence of overhead on LTE performance should not be neglected. The parametrized and dynamic model quantifies to what extent throughput and capacity are modified by overhead—under a combination of system configurations and services, and may provide these performance metrics, throughput and capacity, as inputs to planning, dimensioning and optimization specialized tools.     

On the theory of linear surface acoustic wave devices

Stationary distribution of the efficient wave sources excited by interdigital transducers (IDT) of arbitrary configuration has been derived on the basis of reciprocity of the generation and reception of surface acoustic waves (SAW) in piezoelectrics. It was shown that these sources represent a linear combination of two stationary distributions on the surface S between the piezoelectric and IDT: the distribution of the normal component of SAW flux density formed in the mode of SAW reception by the short-circuit IDT and the field distribution of the SAW electric potential at the same surface S but in the absence of piezoelectric. The Y-parameter matrix of the device was obtained as an electric M×2-terminal network containing M IDT. It was shown that the SAW transfer admittances and SAW radiation admittances contained additional frequency-dependent multipliers. The characteristics take into account the smoothed diffraction, spatial dispersion in IDT, reflections from IDT, and electric loads of the device.     

Temperature dependence of the optical energy gap for the CdS<Subscript>x</Subscript>Se<Subscript>1−<Emphasis Type="Italic">x</Emphasis></Subscript> quantum dots

A temperature dependence of the optical energy gap E _g (T) for the CdS_xSe_1?x quantum dots synthesized in a borosilicate glass matrix was investigated in the range of 4.2–500 K. It was demonstrated that this dependence reproduced the dependence E _g (T) for bulk crystals and is described by the Varshni formula for \(\bar r > a_B \) over the entire temperature range. Here, \(\bar r\) is the average dot radius, and a_B is the Bohr radius for the exciton in a bulk crystal. With the transition to quantum dots with \(\bar r > a_B \), a decrease in the thermal coefficient of the band gap and a deviation from the Varshni dependence were observed in the temperature range of 4.2–100 K. The specific features observed are explainable by a decrease in the resulting macroscopic potential of the electron-phonon interaction and by modification of the vibration spectrum for dots as their volume decreases.     

Influence of the doping type and level on the morphology of porous Si formed by galvanic etching

The formation of porous silicon (por-Si) layers by the galvanic etching of single-crystal Si samples (doped with boron or phosphorus) in an HF/C₂H₅OH/H₂O₂ solution is investigated. The por-Si layers are analyzed by the capillary condensation of nitrogen and scanning electron microscopy (SEM). The dependences of the morphological characteristics of por-Si (pore diameter, specific surface area, pore volume, and thickness of the pore walls), which determine the por-Si combustion kinetics, on the dopant type and initial wafer resistivity are established.     

Self-Action of Intense Millimeter Waves in Waveguides with Integrated P-I-N Structures

The nonlinear interaction of high power millimeter (mm) electromagnetic waves with silicon integral p-i-n structures placed in a metal waveguide is theoretically investigated. The level of double injection of charge carriers due to detection of high intensity millimeter wave electric field in p-i-n structures is estimated. A mathematical model of the mutual influence of electromagnetic waves and injected charge carriers in the active region of p-i-n structures is formulated. A numerical solution of the nonlinear Helmholtz equation supplemented by proper boundary conditions on the active region boundary is obtained. The effect of high-power electromagnetic waves leads to an excessive injection of carriers into the active region of the semiconductor between p⁺-i, n⁺-i injection junctions and redistribution of the electric field in the structure. The reflection and transmission coefficients vary rapidly with the change in the input amplitude of the electromagnetic wave. This leads to bistability of these coefficients. The bistability is more pronounced in the low-frequency part of the mm range.     

Modification of the properties of porous silicon on adsorption of iodine molecules

Infrared spectroscopy and electron spin resonance measurements are used to study the properties of porous silicon layers on adsorption of the I₂ iodine molecules. The layers are formed on the p-an n-Si single-crystal wafers. It is established that, in the atmosphere of I₂ molecules, the charge-carrier concentration in the layers produced on the p-type wafers can be noticeably increased: the concentration of holes can attain values on the order of ～10¹⁸?10¹⁹ cm^?3. In porous silicon layers formed on the n-type wafers, the adsorption-induced inversion of the type of charge carriers and the partial substitution of silicon-hydrogen bonds by silicon-iodine bonds are observed. A decrease in the concentration of surface paramagnetic defects, P _b centers, is observed in the samples with adsorbed iodine. The experimental data are interpreted in the context of the model in which it is assumed that both deep and shallow acceptor states are formed at the surface of silicon nanocrystals upon the adsorption of I₂ molecules.     

A Faddeev Systolic Array for EKF-SLAM and its Arithmetic Data Representation Impact on FPGA

The Extended Kalman Filter (EKF) computation is a core task for the simultaneous localization and mapping (SLAM) problem in autonomous mobile robots. The SLAM problem involves operations over high dimension data sets, requiring high throughput and performance, given the real-time nature of the robotics, control-decision algorithm this task is a part of. The lightweight and power restricted computing environments in mobile robotics requires customized processing systems such as Field-Programmable Gate Arrays (FPGAs). This work presents an arithmetic precision analysis and a Faddeev algorithm to calculate the Schur’s Complement hardware architecture implementation for the EKF-SLAM using a Systolic Array (SA). While it is widely believed that fixed-point implementations of arithmetic operations lead to area and performance benefits on FPGAs, the results in this article reveal that each Processing Element (PE) in the SA consumes 25% more logic and about 30% more register resources for the fixed-point 13.23 representation than if using the IEEE-754 single precision floating-point format. In addition, for FPGA devices with hardware support for key components of floating-point computations, a single PE floating-point implementation can achieve a maximum frequency up to 50% higher than a corresponding fixed-point implementation for the same relative numeric errors.     

Surface states on the <Emphasis Type="Italic">n</Emphasis>-InN-electrolyte interface

Dependences of differential capacitance of the electrolyte-n-InN (0001) contact on the bias voltage are studied. Their analysis of the basis of a model similar to a model of the MIS structure shows that the energy spectrum of surface states of InN above the conduction band bottom can be represented by two, relatively narrow, bands of deep levels described by the Gaussian distribution. Parameters of these bands are as follows: the average energy counted from the conduction band bottom, ΔE ₁ ≈ 0.15 eV and ΔE ₂ ≈ 0.9 eV; and the mean-square deviation, ΔE ₁ ≈ 0.15–0.25 eV and ΔE ₂ ≈ 0.05–0.1 eV. The total density of states in the bands are (1–2.5) × 10¹² and (0.2–4) × 10¹² cm^–2.     

About solution of problem of construction ship radar of higher precision class for short-range navigation purpose

It is shown, that ship radars of S and X wave bands do not satisfy navigation safety conditions in close, coastal and internal waters, because of insufficient efficiency of precision parameters definition, and, consequently, because of small information capacity. Small value of probe pulse duration (about 70 ns) application allows to obtain high distance resolution, but it does not provide high azimuth resolution, because of some reasons. In this paper we consider one method of objects precision parameters improve by means of Ka or W wave bands radars application.     

Electron mobility in the inversion layers of fully depleted SOI films

The dependences of the electron mobility μ_eff in the inversion layers of fully depleted double–gate silicon-on-insulator (SOI) metal–oxide–semiconductor (MOS) transistors on the density N _e of induced charge carriers and temperature T are investigated at different states of the SOI film (inversion–accumulation) from the side of one of the gates. It is shown that at a high density of induced charge carriers of N _e > 6 × 10¹² cm^–2 the μeff(T) dependences allow the components of mobility μ_eff that are related to scattering at surface phonons and from the film/insulator surface roughness to be distinguished. The μ_eff(N _e ) dependences can be approximated by the power functions μ_eff(N _e) ∝ N _e ^?n . The exponents n in the dependences and the dominant mechanisms of scattering of electrons induced near the interface between the SOI film and buried oxide are determined for different N _e ranges and film states from the surface side.     

Performance characteristics of <Emphasis Type="Italic">p</Emphasis>-channel FinFETs with varied Si-fin extension lengths for source and drain contacts

The length of Source/Drain (S/D) extension (L_SDE) of nano-node p-channel FinFETs (pFinFETs) on SOI wafer influencing the device performance is exposed, especially in drive current and gate/S/D leakage. In observation, the longer L_SDEpFinFET provides a larger series resistance and degrades the drive current (I_DS), but the isolation capability between the S/D contacts and the gate electrode is increased. The shorter L_SDE plus the shorter channel length demonstrates a higher trans-conductance (G _m ) contributing to a higher drive current. Moreover, the subthreshold swing (S.S.) at longer channel length and longer L_SDE represents a higher value indicating the higher amount of the interface states which possibly deteriorate the channel mobility causing the lower drive current.     

The mechanism of backscattering of electromagnetic waves by a pine forest in the meter wave band

A synthetic-aperture multifrequency polarimetric radar complex is used for experiments on studying the features of reflection of electromagnetic waves from pine forests. It is found that the intensity of the reflected power is proportional to the biometric characteristics of the forests. The hydrology and spatial fluctuations of the dielectric characteristics of the upper layer of the forest soil are studied at different depths. It is shown that spatial fluctuations of the soil’s dielectric characteristics can be used to explain the features of reflection from the forest. Reflection from a forest in the meter wave band at horizontal (HH) and vertical (VV) polarizations is simulated. It is found that mechanisms of reflection from the forest may be different for the HH and VV polarizations. Experimental data confirming the presence of specific effects in the radar images obtained in the meter wave band at the HH and VV polarizations are presented.