Improve Indoor Position Localization Based on FFNN with PSO Optimization Algorithm

Abstract

In this search, a calculations of Indoor Positioning System (IPS) are applied for the chosen building on one floor. Four Access Points (APs) are installed in the tested area and are specified their locations by using special site survey to insure full coverage area of the case study. The database is created for a selected Reference Point (RP) in a specific area of the work which contained the Received Signals Strength (RSSs) collected from all of directions for each RP, those values are recorded by using a Package of Net Surveyor 0.2. Four layers of NN with back propagation learning algorithm is utilized to act as a localization estimation of user's location. Proposed algorithm is improved by selecting the best number of neurons in each hidden layer and optimal value of learning rate by using the Particle Swarm Optimization (PSO) algorithm. The results showed that the proposed algorithm led to a high degree of accuracy with a very short implementation time.