We present a three-dimensional finite difference time domain (FDTD) method on graphics processing unit (GPU) for plasmonics applications. For the simulation of plasmonics devices, the Lorentz-Drude (LD) dispersive model is incorporated into Maxwell equations, while the auxiliary differential equation (ADE) technique is applied to the LD model. Our numerical experiments based on typical domain sizes as well as plasmonics environment demonstrate that our implementation of the FDTD method on GPU offers significant speed up as compared to the traditional CPU implementations.

A finite element-boundary integral-domain decomposition method is presented for analyzing electromagnetic scattering problems involving multiple three-dimensional cavities. Specifically, the edge-based finite element method is applied inside each cavity to derive a linear system of equations associated with unknown fields. The boundary integral equation is then applied on the apertures of all the cavities to truncate the computational domain and to connect the matrix subsystem generated from each cavity. With the help of an iterative domain decomposition method, the coupling system of equations is reduced to a small one which only includes the unknowns on the apertures. To further reduce computational burdens, the multilevel fast multipole algorithm is adopted to solve the reduced system. The numerical results for the near and far fields of several selected multi-cavity problems are presented to demonstrate the validity and capability of the proposed method.

In this article the influence of both dispersion and losses on waveguides with metamaterials is investigated. The analysis is focused on surface waveguides (planar interfaces and grounded slabs) containing either double-negative (DNG) or chiral metamaterials. The main goal is to show how the combined effect of material dispersion and losses with the structural dispersion affect the solutions of the modal equations. It is shown that this interplay is essential to obtain a correct modal analysis of these waveguides. Namely, the overall behavior can qualitatively change - so that it is not possible to state that the corresponding lossy case - even when a very small amount of losses is introduced - can be interpreted as a small perturbation of the lossless case.

Phase-retrieval from measured phaseless field data is of interest for various applications including electromagnetic dosimetry, electromagnetic compatibility investigations, near-field to far-field transformations and antenna diagnostics. In this study two phaseretrieval methods are compared. The first method, the adjoint field method, employs a gradient-based optimization algorithm based on the adjoint fields. The second method, the phase angle gradient method, uses an optimization algorithm based on the phase angle gradients of a functional. The methods are tested with numerical test cases and the phase angle gradient method is found to retrieve the phase with the best accuracy. Moreover it gives results that agree well with correct phase.

In this letter, a modified broadband 90° phase shifter is proposed. By using a dentate microstrip and a patterned ground plane, an extremely tight coupling can be obtained, and consequently a constant phase shift over a wide bandwidth can be achieved. To verify the proposed idea, a topology is implemented, the measured results of a phase difference of 90 ± 5° in 79.5% bandwidth, better than 10 dB return loss across the whole operating band, are also given. The measurement results agree well with the full-wave electromagnetic simulated responses.

In this paper, we present a frequency tunable antenna suitable for Digital Video Broadcasting Handheld applications. Due to the narrow operating frequency band of the antenna derived from a monopole coupled loop antenna, a tuning system has been proposed to sweep the narrow band on a large frequency range in order to cover the DVB-H frequency bandwidth [470-702] MHz. We provide results of this antenna mounted on a circuit board simulating a terminal handset.

A compact ultrawideband (UWB) antenna with 3.5/5.2GHz band-notched characteristics is proposed. The proposed antenna is composed of a half elliptic annulus radiation element fed by microstrip line and a step-shaped ground plane with truncated corners. By inserting closed-looped resonating structure onto the radiation patch and connecting open-looped resonator on the back side with patch via metallic hole, the dual notched frequency bands are achieved. The numerical and experimental results exhibit a wide impedance bandwidth ranging from 3.08 to 11GHz with the dual notched bands around 3.5 and 5.2GHz. Additionally, nearly omni-directional radiation patterns, moderate gain, and small group delay variation are also obtained.

Photonic wire lasers are compact light sources that are fabricated in high-index contrast waveguides (with typical widths of a few hundreds of nanometers). Because of their small footprints, they may become a basic laser component in future-generation of optical integrated circuits. Owing to having low optical volume by design, photonic wire lasers can only produce low output power that may not be adequate in many applications. A solution to this problem is to coherently combine the output power of different photonic wire lasers to produce larger output power. In this article, we analyze different ways to combine light coming out from photonic wire lasers and couple the combined power into a single-mode waveguide.

The radiated electromagnetic (EM) fields from a rotating current-carrying circular cylinder were numerically simulated in two dimensions using the method of characteristics (MOC), and the numerical results were presented in this paper. To overcome the difficulty of the grid cell distortion caused by the rotating cylinder, the passing center swing back grids (PCSBG) technique is employed in collaboration of MOC in a modified O-type grid system. In order to have clear demonstration of radiated EM fields, the circular cylinder is set to be evenly divided in radial direction into an even number of slices that are made of perfect electric conductor (PEC) and non electric non magnetic material, alternatively. The surface current is assumed to have a Gaussian profile and to flow uniformly along the axial direction on the PEC surface. The radiated electric and magnetic fields around the cylinder were recorded as functions of time and reported along with the corresponding spectra which were obtained through proper Fourier transformation. Several field distributions over the whole computational space are also given.

A new L-shaped chiral structure working in microwave and optical frequency bands has been designed and simulated. The circular dichroism, ellipticity angle, polarization azimuth rotation angle, and effective parameters of this structure, including relative permittivity, relative permeability, chiral parameter and refractive index, are retrieved from simulated transmission and reflection spectra. The results show that the exceptionally strong optical activity is found for the L-shaped chiral structure. Because of the large chiral parameter of this structure, negative refractive index of one circularly polarized wave can be obtained without simultaneously negative permittivity and negative permeability.

An equivalent circuit, made of the chain connection of a number of T-type twoport networks, is proposed for the very accurate representation of the frequency-domain behavior of radially inhomogeneous solitary cylindrical structures, the individual two-port networks being made of frequency-independent R, L and C lumped elements. The accuracy of the model is dictated by the number of two-port networks, a number that increases with the frequency. The equivalent circuit approach is validated with the help of an application example concerning a special type of inhomogeneous tubular structures where exact closedform field solutions do exist.

The diurnal and seasonal variation of surface refractivity over Nigeria was studied using four years in-situ meteorological data from eight location over Nigeria. At all the stations studied, it was observed that the diurnal refractivity variation was caused majorly by the dry term in the rainy season and the wet term is the major cause of refractivity variation in dry season except Sokoto and Jos. In Sokoto the result was found to be opposite and it is attributed to the fact that in dry season the humidity is almost close to zero while in rainy season the pressure seems to be almost constant but the temperature fluctuates rapidly and consequently the humidity. The variation pattern in Jos is as observed because of the altitude (~1000 m above sea level). At this altitude pressure variation seems to be insignificant. The result also show that the surface refractivity generally have higher value during rainy season than dry season at all location studied. The result also show that the value of surface refractivity increases from arid region in the north to the coastal area in south. The result also show that the diurnal refractivity variation is basically a function of local meteorology and while seasonal variation is caused follows the climatic condition.

A novel strategy for topside ionosphere sounder based on spaceborne Multiple-Input Multiple-Output (MIMO) radar is proposed, which takes advantage of frequency division and code division (FDCD) as a substitution for swept-frequency regime employed by the current ionosphere explorers, e.g., TOPside Automated Sounder (TOPAS). The azimuth resolution can be improved by 153 times compared with TOPAS by means of frequency division, producing two-dimensional electron density images. The signal-to-noise ratio (SNR) can be enhanced and complete orthogonality among channels at different frequencies can be achieved by code division, which uses Complete Complementary Sequence (CC-S) as phase coding waveform. The simulation results show that root mean square (RMS) of normalized electron density measurements error of novel ionosphere sounder is as low as 1.7%.

With recent advances in both algorithm and component technologies, through-the-wall sensing and imaging is emerging as an affordable sensor technology in civilian and military settings. One of the primary objectives of through-the-wall radar imaging (TWRI) systems is to detect and identify targets of interest, such as humans and cache of weapons, enclosed in building structures. Effective approaches that achieve proper target radar cross section (RCS) registration behind walls must exploit a detailed understanding of the radar phenomenology, in general, and more specifically, knowledge of the expected strength of the radar return from targets of interest. In this paper, we investigate the effects of various wall types on the received power of the target return through the use of a combined measurement and electromagnetic modeling approach. The RCS of material-exact rifle and human models are investigated in free-space using numerical electromagnetic modeling tools. A modified radar range equation, which analytically accounts for the wall effects, including multiple reflections within a given homogeneous or layered wall, is then employed in conjunction with wideband measured parameters of various common wall types, to estimate the received power versus frequency from numerically modeled aforementioned targets of interest. The proposed technique is, in principle, applicable to both bistatic and monostatic operations. The results for various wall types, including drywall, brick, solid concrete and cinder block, under both wet and dry conditions, are presented.

This paper proposes a new approach for efficiently determining the unwanted interfering samples in the reference window, for the ordered statistics constant false alarm rate detector, based on the application of the information theoretic criteria principle. The proposed processor termed as Forward Automatic Order Selection Ordered Statistics Detector (FAOSOSD) does not require any prior information about the number of interfering targets. The proposed design aims to improve the Ordered Statistics Constant False Alarm Rate detector performance under severe interference situations. The number of interfering targets is obtained by minimizing the information theoretic criteria. Simulation results that illustrate the performance of the proposed method versus the classical OS-CFAR, the AND-CFAR and the OR-CFAR detectors are presented and discussed.

A dual-mode dual-band bandpass filter is designed using a single stub-loaded slot ring resonator. This resonator is coupled to the two external feed lines at two positions spaced at 135° along the slot ring through a pair of microstrip-slotline Tjunctions. With a proper choice of the degree of external coupling, the first-order degenerate modes are split to make up the first passband with two transmission poles. The second passband is realized by the second-order degenerate modes, which are stimulated by symmetrically attaching four identical stubs along the slot ring. The center frequency ratio of the two operating passbands is controlled by the nature and strength of the external coupling, which is determined by the characteristics of the microstrip opencircuited stubs. Finally, a dual-band filter with center frequencies at 2.4 and 5.2 GHz is designed and fabricated. Measured results verify the design principle and predicted dual-passband performance. Benefiting from an additional transmission zero brought by the transitions, the upper stopband is expanded up to 12.75 GHz with at least 13 dB of rejection.

A planar monopole antenna with reconfigurable radiation patterns is demonstrated. The radiation pattern reconfigurability is realized straightforwardly with an employment of a detached magnetodielectric slab placed in the vicinity of the antenna structure. It is shown the radiation patterns can be easily reconfigured through the adjustment of the spacing between the slab and the antenna structure. Technically, the radiated fields are redistributed owing to the inclusion of the magnetodielectric slab, which is of high permittivity, as well as permeability. As a result, the planar monopole gain with the slab is increased up to 4 dBi while the antenna resonant frequency remains almost unchanged.

In this paper, a novel approach has been suggested to obtain an improved spurious-free window for dielectric resonator in microwave integrated circuit environment. In microwave integrated circuit environment, the dielectric resonator placed on a thin dielectric substrate gets located asymmetrically with respect to its shielding enclosure. A reduced separation in frequencies (mode separation) is one of a consequence of this asymmetry that may become a cause of spurious modes. This adverse influence of asymmetry is sought to be compensated by proposing a multi-layer multi-permittivity dielectric resonator structure with several layers of differing permittivity. The suggested approach takes advantage of the fact that the mode separation of a dielectric resonator configuration can be correlated to relevant resonance mode fields. By perturbing the resonance mode fields through the suggested multi-layer multi-permittivity approach, the adverse influence of asymmetry is found to reduce considerably over a comparative conventional ring dielectric resonator in microwave integrated circuit configuration. Still more improvement in mode separation are shown when the shape of the multi-layer multi-permittivity ring dielectric resonator is further modified, suggesting a scope for optimization in present approach.

A microstrip bandpass filter is presented based on Complementary Split Ring Resonators (CSRRs) and a pair of open-loop resonators that has a single pair of transmission zeros at finite frequencies that causes an improvement at skirt response. An equivalent circuit is introduced to make analysis and optimization faster. Finally a filter is designed using the proposed cell and the simulation results with both equivalent model and full wave analysis are in very good agreement. The filter was fabricated and the measurement result was also in good agreement with simulation results. Besides, the size of the designed filter is very small and it occupies an area less than 0.23λg × 0.16λg, where λg is the guided wavelength at the midband frequency.

A planar sleeve dipole array antenna is analyzed and successfully implemented. The proposed antenna is designed for operation at 1800/1900 MHz band of basic station applications. To achieve sufficient bandwidth for the requirement of the PCS 1800 MHz band (1710-1880 MHz) and 1900 MHz band (1880-1930 MHz) for DECT (Digital Enhanced Cordless Telecommunications), the proposed antenna comprises of a 1 × 5 coplanar back-to-back sleeve dipole elements, we adopt the microstrip line to balanced transmission line feeding technique in this design. This structure is easily constructed by printing on both sides of a dielectric (FR4) substrate. The measured -10 dB return loss (VSWR 2 : 1) impedance bandwidth is around 13.2% (1690-1930 MHz). A reflector is put behind the dipole array to obtain directional radiation and high gain, and the measured antenna gain for operating frequencies across the 1800/1900 MHz band is about 7.2-9.1 dBi. The measured results of radiation efficiency, radiation pattern, antenna gain and return loss show the sleeve dipole array antenna has a good performance.