Quick Description

Programmers : Zakaria Abdelmoiz DAHI.

About: This repositiory contains the implementation of the quantum-inspired genetic algorithm devised in [1] for solving the binary antenna placement using a conitnuous swarm algorithm. We use instances of 149 to 1000 antennas (representing the city of Malga, Spain) and three types of antennas (Omnidirectional, directional and Squared). Evtually, I made an option for testing a fourth strategy where all types of antennas can be used.

• [1] Z.A. DAHI, C. Mezioud, A. Draa, A quantum-inspired genetic algorithm for solving the antenna positioning problem, Swarm and Evolutionary Computation, Volume 31, 2016, Pages 24-63, ISSN 2210-6502, https://doi.org/10.1016/j.swevo.2016.06.003.

How 📗

• Depending on the variant you want to execute you just need to navigate to the corresponding foldr: GGA for the generational genetic algorithm, PBIL for the population-based incremental learning and QIGA for the quantum-inspired genetic algorithm.
• Once you have navigated to the folder of the corresponding variant, you just need to execute the file main.m.

Folders Hiearchy 📂

• GGA: This folder contains the generational genetic algorithm.

• QIGA: This folder contains quantum-inspired genetic algorithm.

• PBIL: This folder contains the code of the Population-Based Incremental Learning.

• Results:

  • Graphical: the results will be automatically stored as gif figures.
  • Nurmerical: the results will be stored as Excel files with name as instance_shape.xls, where instance is the size of the benchmarks (i.e. the number of candidate Antennas) and shape is the shape of the antenna being using. The size ranges from 149 to 1000 candidate antennas, while for the shapes we use omnidirectional, directional and rectangular.

Demo 🎥

• Please refer to the original paper HERE for more detailed results and discussions.