Design of Custom-Made Fractal Multi-Band Antennas Using ANN-PSO

We present a design procedure for making custom-made fractal antennas using artificial neural networks (ANN) and the Particle-Swarm Optimization (PSO) technique. The role of the artificial neural network is to form a mapping between the design parameters of the fractal antenna and its operational frequencies. The role of Particle-Swarm Optimization is to find the shape of the antenna for the required user-defined frequencies, using the previously trained artificial neural network. Sierpinski gasket and Koch monopole antennas were taken as the candidate antennas, and the effectiveness of the developed approach was confirmed by simulation and experimental results. Antenna shapes at various user-defined frequencies were designed and tested.