Real-time RSS-based positioning system using neural network algorithm

Locating services have come under the spotlight in recent years various applications. However, locating methods that use received signal strength low accuracy due to fluctuations. For this purpose, we present a Wi-Fi based system using artificial neural network enhance positioning process performances. We optimized Levenberg Marquardt algorithm propose better configuration of multi-layer time-delay perception network. achieved an average error 10.3 centimeters with grid 0.4 meter four tests. Yet, instability RSS-based systems limitation resolution finesse depends on size.