Analysis of the Machine Learning Methods for Optimizing the Wireless Access Point Position Indoors

The paper considers introduction of the machine learning methods to optimize the wireless access point position in the indoor environment. It compares various methods, including gradient descent, genetic algorithms, particle swarm optimization, and reinforcement learning, and evaluates their suitability in achieving the optimal coverage. The paper presents results of an experiment conducted to test the practical functionality of these methods, highlights strengths and limitations of each of them and focuses on their ability to maximize the signal strength. The paper concludes that combination of the gradient descent and genetic algorithms could offer the best approach and ensure efficient signal distribution and network scalability in the complex environments.