Deterministic methods in indoor-localization systems based on the received signal strength (RSS) almost utilize the average value of the RSS, such as the k-nearest neighbor (KNN) algorithm. However, the distribution of RSS is not always normal Gaussian in the real complex indoor environment so the average value may not represent the location well. To solve this problem, we present a novel algorithm, named as probabilistic KNN (p-KNN) algorithm. The algorithm uses the probability of RSS in the Radio-map as a weighting to calculate the Euclidean distance, and it filters the RSS value whose probability is less than 3%.
At the same time, we propose a new application environment called as passive indoor-localization scenario. In this scenario, the access point (AP) collects the RSS when the mobile terminal (MT) is not connecting to the AP. Experiment and results analysis for different k values show that p-KNN algorithm is feasible and effective in passive indoor-localization scenario. Finally, comparing to the KNN algorithm, p-KNN algorithm can achieve a better average location accuracy.Index Terms-Received signal strength(RSS); Passive indoorlocalization; K-nearest neighbor (KNN).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.