The use of antennas for vehicle applications is growing very rapidly due to the development of
modern wireless communication technology and service. Currently, an automobile antenna design
using the Simple Genetic Algorithm (SGA) has been introduced. In this model, the SGA computation
tool attempts to obtain the best design based on a single cost function. The automobile antenna
design is a multi-objective problem. The different objectives are combined into a single cost
function, each with a weight value. The results of the optimization procedure depend strongly on
these weights, and thus, the designer must properly choose each weight value to get the desired
optimum. Also, all the weight values must then be changed, and the entire optimization procedure
must be repeated whenever the designer wants to change any single objective goal. This
dissertation presents the development and application of the Nondominated Sorting Genetic
Algorithm (NSGA) to design new automobile conformal antennas. The NSGA can find a set of Pareto-
optimal solutions, instead of finding a single optimal solution. In multi-objective optimization
problems, one may not find a single best solution. There may be many solutions, which are
considered better with respect to all objectives. The Pareto-optimum solutions are a set of
compromise solutions based on a comparison with each objective. The NSGA searches the Pareto-
optimal solutions by using a fitness assignment process and a sharing process. A set of Pareto-
Optimum automobile conformal antenna geometries for FM radio and GPS/SDARS systems using the NSGA
is produced. The use of antennas for vehicle applications is growing very rapidly due to the
development of modern wireless communication technology and service. Currently, an automobile
antenna design using the Simple Genetic Algorithm (SGA) has been introduced. In this model, the
SGA computation tool attempts to obtain the best design based on a single cost function. The
automobile antenna design is a multi-objective problem. The different objectives are combined into
a single cost function, each with a weight value. The results of the optimization procedure depend
strongly on these weights, and thus, the designer must properly choose each weight value to get
the desired optimum. Also, all the weight values must then be changed, and the entire optimization
procedure must be repeated whenever the designer wants to change any single objective goal. This
dissertation presents the development and application of the Nondominated Sorting Genetic
Algorithm (NSGA) to design new automobile conformal antennas. The NSGA can find a set of Pareto-
optimal solutions, instead of finding a single optimal solution. In multi-objective optimization
problems, one may not find a single best solution. There may be many solutions, which are
considered better with respect to all objectives. The Pareto-optimum solutions are a set of
compromise solutions based on a comparison with each objective. The NSGA searches the Pareto-
optimal solutions by using a fitness assignment process and a sharing process. A set of Pareto-
Optimum automobile conformal antenna geometries for FM radio and GPS/SDARS systems using the NSGA
is produced.