Fast location of high-order modes for boxed microstrip with finite strip thickness

Many analyses of boxed microstrip discontinuities require the location of large numbers of high-order modes. Using Sturm-Liouville theory, this letter derives a relationship between microstrip modes and slab-loaded guide modes, leading to an efficient mode location algorithm for microstrip, including microstrip with finite strip thickness.<>     

Three-Dimensional FDTD Simulation of Micro-Pillar Microcavity Geometries Suitable for Efficient Single-Photon Sources

We present the results of calculations of the microcavity mode structure of distributed-Bragg-reflector (DBR) micro-pillar microcavities of group III-V semiconductor materials. These structures are suitable for making single photon sources when a single quantum dot is located at the center of a wavelength scale cavity. The 3-D finite difference time domain (FDTD) method is our primary simulation tool and results are validated against semi-analytic models. We show that high light extraction efficiencies can be achieved (>90%) limited by sidewall scattering and leakage. Using radial trench DBR microcavities or 2-D photonic crystal structures, we can further suppress sidewall emission, however, light is then redirected into other leaky modes     

Characterisation of microstrip open-end with rectangular and trapezoidal cross-section

The vast majority of published analyses of microstrip discontinuities make the, often unrealistic, approximation of infinitesimally thin strips. Results obtained using the finite difference time domain (FDTD) method are presented for the open end discontinuity in microstrip with various cross-sections. These demonstrate that the cross-section significantly influences the parasitic effects.<>     

A stable subgridding algorithm and its application to eigenvalueproblems

In this paper, a new and stable subgridding algorithm is proposed for three-dimensional problems which provides subgridding in both space and time. The concept of an equivalent-circuit representation and a novel leapfrog time integration scheme is used to ensure that the algorithm is stable and efficient. Practical applications of this algorithm in the characterization of arbitrarily filled dielectric resonators are reported     

An analytical and numerical analysis of several locally conformalFDTD schemes

The virtues of the finite-difference time-domain (FDTD) method for the electromagnetic analysis of arbitrary complex metal and dielectric structures are well known. Almost equally well known are the difficulties encountered by the technique when the material boundaries do not coincide with the Cartesian mesh. Until recently, there were few alternatives to the simple, but inaccurate, staircase approximation for these cases. However, over the past few years, there have been several solutions proposed, which maintain the simplicity and efficiency of the FDTD method while providing an accurate treatment of curved, offset, or sloping metallic boundaries. In this paper, analytical and numerical comparisons are presented and a clear recommended method is shown to emerge     

A new technique for the stable incorporation of static fieldsolutions in the FDTD method for the analysis of thin wires and narrowstrips

The behavior of the fields around many common objects (e.g., wires, slots, and strips) converges to known static solutions. Incorporation of this a priori knowledge of the fields into the finite-difference time-domain (FDTD) algorithm provides one method for obtaining a more efficient characterization of these structures. Various methods of achieving this have been attempted; however, most have resulted in unstable algorithms. Recent investigations into the stability of FDTD have yielded criteria for stability, and this contribution for the first time links these criteria to a general finite-element formulation of the method. It is shown that the finite-element formulation provides a means by which FDTD may be generalized to include whatever a priori knowledge of the field is available, without compromising stability. Example results are presented for extremely narrow microstrip lines and wires     

The treatment of thin wires in the FDTD method using a weighted residuals approach

In this contribution, the problem of accurately representing thin wires within the finite-difference time-domain (FDTD) mesh is addressed by means of a method based on the weighted residual (WR) interpretation of the FDTD algorithm. Results for wire dipoles and wire transmission lines, obtained using the proposed method, are presented and compared to those obtained using existing techniques. It is shown that the proposed method yields results which are more accurate and are less dependent on the choice of cell size than other approaches and, in addition, lends itself well to being extended for more complicated structures. Details of the calculation of the update equations are given.     

Reducing the computational overhead of the near-field transformthrough system identification

The system identification technique is applied to the output of a time-domain near to far-field transform employed with the FDTD algorithm. The technique is used to characterise the far field of a microstrip antenna, the accuracy of the results is evaluated, and the computational savings and overheads involved are discussed     

The analysis of medium-sized arrays of complex elements using acombination of FDTD and reaction matching

The analysis of medium-sized arrays of complex antenna elements by means of a full-wave technique often requires impractical amounts of computer power. Nevertheless, it is essential that all the mutual couplings between elements are taken into account. A technique is presented in which the individual element is characterized using the FDTD method and, using the information this provides, the behavior of the complete array is predicted using a method based on reaction matching. Results using this method are compared to measurement and to results obtained using a complete full-wave analysis for three- and five-element arrays of printed dipoles. The error introduced by the approximation is shown to be small in most cases. For arrays of between 10 and 50 elements, savings in computer time of several orders of magnitude can be achieved and, in addition, changes in array geometry do not always necessitate all the results being recalculated     

Practical method for the determination of time step innon-orthogonal FDTD

The non-orthogonal FDTD algorithm is well-suited to the modelling of curved structures, although users of the technique frequently have difficulty determining an appropriate time step for the algorithm. A new method which is shown to be both a practical and reliable technique for determining the time step is presented