Neelakantam V. Venkatarayalu

Optimum design of Yagi-Uda antennas using computational intelligence

Optimization of Yagi-Uda antennas is a challenging design problem, since the antenna characteristics such as gain, input impedance, maximum sidelobe level etc., are known to be extremely sensitive to the design variables viz., element lengths and their spacings. This corresponds to a highly nonlinear and multimodal function space with functional and slope discontinuities that limit the use of conventional gradient based optimization approaches. Although, stochastic, zeroth-order methods like genetic algorithm and evolutionary algorithm are attractive choices for such classes of problems, their successful application requires scaling and aggregating parameters to handle constraints and objectives that may not be easy to provide. In this paper, we introduce a stochastic, zeroth-order optimization algorithm that handles constraints and objectives separately via Pareto ranking that eliminates the problem of scaling and aggregation. The algorithm is based on principles of learning and is embedded with three key learning strategies that control whom to learn from (i.e., leader identification and leader selection) and what to learn (i.e., information acquisition) in order to better guide the search. The leader identification mechanism partitions the individuals into a set of leaders and a set of followers. The followers interact with the leader and move toward the better performing leaders in search for better solutions. As the algorithm does not require parameters for scaling or aggregation, it provides the designer the true flexibility that is necessary to handle various forms of the design problem effectively and at a computational cost that is comparable to existing stochastic optimization methods. Results of three single objective antenna design examples (a four-element, a 15-element and a fixed boom length 22-element design) are presented and compared with published results to illustrate the behavior of the proposed algorithm and highlight its benefits in solving a wide variety of antenna design problems. A new set of results are presented for a multiobjective formulation of the design problem.

A Stable FDTD Subgridding Method Based on Finite Element Formulation With Hanging Variables

The finite-element time-domain (FETD) method based on the use of hanging variables to generate nested grids is used as an interface between the coarse and fine grids in the FDTD subgridding method. Since the formulation for treating hanging variables is based on a Galerkin-type intergrid boundary operator, the resulting FDTD subgridding algorithm is guaranteed to be stable. Numerical examples such as the computation of resonant modes of a 3-D rectangular resonant cavity and 2-D TEz scattering by a PEC cylinder and a NACA airfoil are presented to verify the stability and accuracy of the proposed subgridding algorithm

General formulation of unconditionally stable ADI-FDTD method in linear dispersive media

The unconditionally stable alternating-direction-implicit-finite-difference time-domain (ADI-FDTD) method is used to model wave propagation in dispersive media. A formulation is presented by introducing the Z-transform method into the ADI-FDTD scheme to handle the frequency-dependent features of the media. This formulation is applicable to arbitrary dispersive media, and can be easily coded. Numerical results are compared to those based on the conventional FDTD method to show the efficiency of the proposed method.

Single and multi-objective design of Yagi-Uda antennas using computational intelligence

Design of Yagi-Uda antennas is a challenging problem since antenna characteristics such as gain, input impedance, maximum sidelobe level etc., are known to be extremely sensitive to the design variables viz., element lengths and their spacings. Although, population-based, stochastic, zero-order methods like genetic algorithm (GA) and evolutionary algorithm (EA) are attractive choices for such classes of problems, their successful application requires a number of additional inputs (e.g. scaling and aggregating factors to deal with constraints and objectives) that is not easy for a designer to provide. We introduce a population-based, stochastic, zero-order optimization algorithm and use it to solve single and multiobjective Yagi Uda design optimization problems. The algorithm is attractive as it is computationally efficient and does not require additional user inputs to model constraints or objectives. One single objective and two multiobjective Yagi Uda design examples are presented. The first example highlights the limitations of using an aggregate objective function in design optimization, while the second and the third examples illustrate the performance of our optimization algorithm for multiobjective problems.

Multilayer dielectric filter design using a multiobjective evolutionary algorithm

Design of multilayer dielectric filters involve the identification of suitable dielectric material and appropriate thicknesses of the layers that best satisfies the desired frequency response for the application. Such problems, like any other practical design optimization problem require simultaneous consideration of multiple objectives and constraints. In this paper, we introduce a multiobjective evolutionary algorithm that is capable of handling unconstrained and constrained, single and multiobjective problems without any restriction on the number and nature of variables, constraints and objectives. The algorithm handles constraints and objectives separately using two fitness measures derived out of nondominance, unlike most of its counterparts which use a single fitness measure. Unlike most evolutionary algorithms where only the good parents participate in mating, our algorithm ensures that all solutions participate in mating, which is useful for exploring highly nonlinear search spaces. The diversity of the solutions is controlled by the partner selection scheme that prefers elites with distant neighbors as mating partners. The results of two multiobjective test problems, three multilayer dielectric filter designs (low-pass, bandpass and stopband) and one variable layer low-pass filter design are presented in this paper to highlight the benefits offered by our algorithm in terms of modeling flexibility, computational efficiency and its ability to arrive at competitive nondominated designs. A comparison of our results with those obtained using a single objective aggregated formulation for the stopband filter design is also presented. We have also compared the performance of our algorithm with nondominated sorting genetic algorithm (NSGA-II) for the low-pass filter design where our results are better.

Suppressing linear time growth in edge element based finite element time domain solution using divergence free constraint equation

Edge elements have been a popular choice of basis functions for the solution of Maxwells equations using the finite element method. When applied in time domain finite elements (Lee, J.-F., 1994), late time linear time growth (Hwang, C.-T. and Wu, R.-B., 1999) is observed. A method to suppress the linear time growth observed in FETD (finite element time domain) solutions is presented. The method is based on constraint equations to restrict the vector finite element solution to be divergence-free. Thus, gradient vector functions in the solution are suppressed. The method is applied to a ridged cavity and the elimination of non-physical low frequency terms is observed.

Application of Hybrid FETD-FDTD Method in the Modeling and Analysis of Antennas

The stable hybrid finite-element time-domain-finite-difference time-domain (FETD-FDTD) method is applied for the numerical modeling and simulation of radiation from antennas. Use of unstructured tetrahedral elements in the modeling of antenna structure enables the application of the hybrid method to accurately model geometrically complex radiators. Traditional FDTD method with anisotropic perfectly matched layer (PML) is used to simulate unbounded media. Pyramidal elements are used in the transition from unstructured tetrahedral elements to structured hexahedral elements of the FDTD grid. The hybrid method is extended by using hierarchical mixed order basis functions in the unstructured region. The finite element formulation incorporates the excitation of antennas using coaxial line or stripline feed with transverse electromagnetic mode (TEM). Application of this method in the modeling of typical wideband antennas along with the results of input reflection coefficient and radiation pattern is presented.

Fast monostatic RCS computation in FEM based solvers using QR decomposition

A technique to reduce computational time in problems involving monostatic RCS computation using the finite element method is presented. In scattering problems, the excitation vectors for different angle of incidence are highly linearly dependent. Using this fact, the excitation matrix assembled over all incident angles can be approximated with a specified tolerance by QR decomposition with column rank that is much less than the number of incident angles of interest. The number of matrix solutions required is reduced to the effective column rank, resulting in significant reduction in computational time. Numerical examples highlighting the effectiveness of the proposed acceleration technique is presented.

Evaluation of IBIS modelling techniques for signal integrity simulations without and with package parasitics

Input/Output Buffer Information Specification (IBIS) models are widely used in signal integrity analysis because of their ability to protect proprietary information and to reduce simulation time when compared to full SPICE simulations. Generation of IBIS models with I/V and V/T data from a full SPICE model of a typical digital buffer without and with package parasitics is investigated in this paper. Several different IBIS model generation strategies to incorporate package effects are validated with the full SPICE model in order to provide a suitable approach. In addition, the accuracy of IBIS simulations in HSPICE and ADS is investigated.

Study on the behaviour and implementation of parent centric crossover within the generalized generation gap model

We report the results of our study on the behaviour and implementation of the parent centric operator (PCX) within the generalized generation gap (G3) model using five test functions of 10, 20 and 50 dimensions. Our study indicates that G3-PCX performs fairly well on most functions, but its performance is not good for highly nonlinear, multidimensional problems (Rastrigin, Ackley, Griewangk). We observed the same behaviour of G3-PCX while designing a 22 element Yagi-Uda Antenna for gain maximization (known to be a highly nonlinear problem). We derived a simple variant G3-PCX-II using a Roulette wheel based parent selection scheme which performs better than G3-PCX on the highly nonlinear multidimensional problems.