作者: Yuki Maekawa , Akira Uchiyama , Hirozumi Yamaguchi , Teruo Higashino
关键词:
摘要: We propose a method to estimate car-level train congestion using Bluetooth RSSI observed by passengers' mobile phones. Our approach employs two-stage algorithm where location of passengers is estimated infer congestion. have learned signals attenuate due bodies, distance and doors between cars through the analysis over 50,000 real samples. Based on this prior knowledge, our designed as Bayesian-based likelihood estimator, robust change both at stations. The positions are useful for personal navigation inside stations information helps determine better strategies taking trains. Through field experiment, we confirmed can 16 with 83% accuracy also 0.82 F-measure value in average.