Application of an Enhanced Genetic Algorithm to Radar System

Abstract: This paper introduces an enhanced Genetic Algorithm (GA) that is faster and less computationally expensive than the standard GA. Four enhancements are introduced here (multiple weighted roulettes, multiple cross over points, multiple mates and utilizing the D4 wavelets). Then the new enhanced system was applied on a dynamic large optimization problem that consists of an array of sixteen radar sensors to predict the Angle of Arrival (AoA) for an approaching object. Results was obtained using the enhanced GA as well as the standard GA. The enhanced GA is able to find the AoA using the least possible number of calculations, which, means it was less expensive computationally and is robust to parameter selections. Special software was developed using Java for this purpose; also, MS-Excel was used to represent the data map as charts.

Keywords: Genetic Algorithm; bio inspired system; evolutionary algorithm; weighted roulette wheel; Daubechies wavelets.