Using genetic algorithms for radar waveform selection

Abstract

Genetic algorithms have proven to be useful tools in optimizing complex problems with large solution spaces. Radar waveform selection is a challenging problem that may benefit from the use of genetic algorithms. Furthermore, advances in the areas of waveform diversity, multistatic radars and knowledgeaided radars are making waveform selection even more challenging. As a design tool we used genetic algorithms to perform waveform selection utilizing the autocorrelation and ambiguity functions in the fitness evaluation. Monostatic, bistatic and multistatic notional examples are presented and early results indicate that genetic algorithms can provide a useful and effective tool in waveform selection for a variety of radar configurations.

Radar Angle of Arrival System Design Optimization Using a Genetic Algorithm

“…In [24], Capraro et al utilized a GA to aid in the selection of optimum waveform configurations. The goal was to find a configuration that minimizes the radar ambiguity function based on some provided specifications.…”

Radar tracking waveform design in continuous space and optimization selection using differential evolution

“…In addition, to date the only tractable solutions have been limited to linear chirp functions for their search space and suffered from sensitivity to differentiation noise [5]. The use of genetic algorithms for waveform design has been investigated in [12]. However, the ambiguity function was used for the fitness function, and therefore optimization was not directly attempting to minimize the PMSE for tracking.…”

“…In addition, the family of waveforms was restricted to bi-phase and poly-phase codes that do not contain nonlinear frequency-modulated (FM) signals that are often of interest in cluttered and wideband environments [4]. The goal of [12] was to find binary coded sequences when…”

“…While this was a great improvement over previous works, the library was limited to grid points chosen in advance, and so potentially valuable waveforms between the grid points could never be chosen. Alternative methods of optimization for waveform design, such as genetic algorithms, have been explored to reduce integrated sidelobe levels and improve the ambiguous free range of radar systems [12]. However, they were employed only over the set of bi-phase and poly-phase fixed frequency and fixed amplitude waveforms.…”

Q-Learning-Based Adaptive Waveform Selection in Cognitive Radar

“…In [6], radar waveform selection algorithms for tracking accelerating targets are considered. In [7], genetic algorithms are used to perform waveform selection utilizing the autocorrelation and ambiguity functions in the fitness evaluation. In [8], Incremental Pruning method is used to solve the problem of adaptive waveform selection for target detection.…”