An Enhanced Technique for Indoor Navigation System Based on WIFI-RSSI

Determining position and route is very important because it helps user get to the destination easier and faster. Nowadays, more and more people move to urban areas and live in complex buildings. Indoor positioning and navigation, therefore, plays an important role for determining position for indoor areas. Anyway, in order to get to the destination, knowing only position is not enough because there are a lot of rooms inside building, for example, airport and shopping center. Thus knowing the route to the destination is also very important so user can reach the destination in time. Indoor positioning focuses on using smartphone to receive Wi-Fi signal due to its convenience and ease of operation. The Wi-Fi Fingerprint based localization with k-Nearest Neighbor (k-NN) algorithm has been commonly used for indoor positioning. For indoor navigation, many researches have used structure of building such as distance space and doors as reference points but this research focus on reference nodes of fingerprint map because reference nodes of fingerprint map are built depending on structure of building. This research is divided into two parts. The first part is to improve accuracy and robustness of positioning by using k-NN algorithm with Particle Filter (PF). The second part is the navigation technique for indoor environment by using Dijkstra's algorithm with reference nodes of fingerprint map to find the shortest route from starting position to the destination. For experimental results, the map of 6th floor of New Main Building (NMB) in Beihang University was used for simulation. In positioning part, the results showed the accuracy of k-NN and PF algorithm by using root mean square error equation to measure errors between real position and estimated position. In navigation part, the results showed time used to calculate the route. Moreover, the results also showed the route between starting position and destination.