Sensitivity of a genetic algorithm solution for a wire antenna geometry

The tolerance of the optimized parameters obtained using a genetic algorithm (GA) is an important consideration in the practicality of the design. Here, the geometry of a vertical loaded monopole (VLM) is obtained using a simple GA. To understand the effect of errors in the geometry, a metric called the range improvement factor is developed, a multiplicative factor for the required distance between two quarter-wavelength monopole antennas, which is achieved by using the superior VLM design. The results of corrupting the VLM geometry with Gaussian distributed geometry errors is investigated and indicates that the original design is on a stable plateau within the solution space.