Efficient Global Optimization (EGO) minimizes expensive cost function evaluations by correlating evaluated parameter
sets and respective solutions to model the optimization space. For optimizations requiring destructive testing or lengthy
simulations, this computational overhead represents a desirable tradeoff. However, the inspection of the predictor space
to determine the next evaluation point can be a time-intensive operation. Although DACE predictor evaluation may be
conducted for limited parameters by exhaustive sampling, this method is not extendable to large dimensions. We apply
EGO here to the 11-dimensional optimization of a wide-band fragmented patch antenna and present an alternative
genetic algorithm approach for selecting the next evaluation point. We compare results achieved with EGO on this
optimization problem to previous results achieved with a genetic algorithm.
Chromosome design has been shown to be a crucial element in developing genetic algorithms which approach global
solutions without premature convergence. The consecutive positioning of parameters with high-correlations and
relevance enhances the creation of genetic building blocks which are likely to persist across recombination to provide
genetic inheritance. Incorporating positional gene relevance is challenging, however, in multi-dimensional design
problems. We present a hybrid chromosome designed for optimizing a fragmented patch antenna which combines
linear and two-dimensional gene representations. We compare previous results obtained with a linear chromosome to
solutions obtained with this new hybrid representation.
This paper describes the application of biologically-inspired algorithms and concepts to the design of wideband antenna
arrays. In particular, we address two specific design problems. The first involves the design of a constrained-feed
network for a Rotman-lens beamformer. We implemented two evolutionary optimization (EO) approaches, namely a
simple genetic algorithm (SGA) and a competent genetic algorithm. We conducted simulations based on experimental
data, which effectively demonstrate that the competent GA outperforms the SGA (i.e., finds a better design solution) as
the objective function becomes less specific and more "general." The second design problem involves the
implementation of polyomino-shaped subarrays for sidelobe suppression of large, wideband planar arrays. We use a
modified screen-saver code to generate random polyomino tilings. A separate code assigns array values to each element
of the tiling (i.e., amplitude, phase, time delay, etc.) and computes the corresponding far-field radiation pattern. In order
to conduct a statistical analysis of pattern characteristics vs. tiling geometry, we needed a way to measure the
"similarity" between two arbitrary tilings to ensure that our sampling of the tiling space was somewhat uniformly
distributed. We ultimately borrowed a concept from neural network theory, which we refer to as the "dot-product
metric," to effectively categorize tilings based on their degree of similarity.
Genetic and evolutionary optimization techniques have been used in military antenna research and design at many levels, ranging from electrically-small antenna element design to broadband applications and array-pattern control. In this paper, we summarize in-house work in these areas, conducted at the Antenna Technology Branch of the Air Force Research Laboratory Sensors Directorate. In particular, we highlight areas where differences in modeling and simulation techniques have proven crucial in avoiding premature convergence and obtaining a valid optimal solution.