Increased Accuracy in Indoor Location based on Neural Networks

Abstract

The use of WiFi is widely used by a large number of devices, including those that make up the Internet of Things (IoT) and Artificial Intelligence (AI) systems. The location problem has been under investigation for a long time. In some cases, the radio signals used to transmit information are also used to make position estimates. However, its use is affected by the constant fluctuation of the signal. It is possible that when estimating the position of a component, it is influenced by obstacles, multipath and signal reflection. Its use improves when spatial localization is carried out, where assets can be traced within an indoor environment. In this work, the relationship of the distance estimation algorithms using RSSI and triangulation is analyzed, and a solution based on Neural Networks is proposed that combines the results of three distance estimation algorithms in order to increase precision.

Keywords

• localización
• triangulación
• distancia
• ann