Generative Artificial Intelligence Using Machine Learning on Wireless Ad Hoc Networks

In this article, we discuss the use of Generative Artificial Intelligence (GenAI) to improve the efficiency and performance of access to wireless points located in various spaces and specific places, which allows interaction with wireless mesh networks and enables the use of mobile devices to access all types of information in internal environments. Furthermore, we propose the use of generative neural networks, which are one of the pillars of GenAI, since they use a methodology from the perspective of Machine Learning that allows analysis of a large amount of data and detection of certain types of patterns that help in the better placement of access points for improved reception and connectivity. Images (heat maps), access point locations, positioning points, and bandwidth are analyzed, allowing new information to be created. On the other hand, to understand and model the general architecture of the wireless Ad-Hoc network, we use two processes that are part of neural networks, such as Multilayer Perceptron (MLP), and the Radial Basis Function (RBF), which is a function of predictors or independent variables or input variables that allows the prediction error in the output variables of the wireless network architecture to be reduced. Using these two processes does help reduce blind spots in those internal places where the wireless signal does not reach, resulting in a signal drop. Improving internal scenarios with wireless Ad-Hoc networks is what is required for better functioning and performance of the network infrastructure.