A Review Paper: Indoor Positioning Systems Using Radio Frequency Technology

In this study, Wi-Fi fingerprinting positioning methods have been analyzed where the received signal strength (RSS) from the different access points (APs) is measured and stored in a database during the offline phase. The stored data is utilized through a machine learning algorithm and used to accurately measure the mobile devices' locations in an indoor environment during the online phase. This method is expected to meet the increasing demand of indoor location-based services. Although, there has been an exponential demand for indoor localization in smart buildings in recent times, the challenges of reduced positioning accuracy, unreliability, and longer response time render indoor location-based systems ineffective. The inefficiency and ineffectiveness are caused by several noise types found inside a building that have an impact on the signal strength in radio channels resulting in the corruption of the RSS values. To overcome these challenges, this paper reviewed different indoor location systems methods and algorithms based on their performance and cost. The analysis differentiates between the fingerprinting and triangulation classes of methods. The improvements in accuracy and response time offered by different computational techniques from recently published literature are delved into. The analysis in this work identifies the superiority of fingerprinting methods and neural network algorithms.