An Improved Weighted K-Nearest Neighbors Algorithm for High Accuracy in Indoor Localization

Pham, Hoang-Anh; Nguyen, Quang-Thien-Tri; Le, Thanh-Van

doi:10.1109/apcc47188.2019.9026503

Cited by 6 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Indoor positioning using Wi-Fi RSS fingerprinting was broadly addressed across literature, most commonly considering k-NN [3], [4] or various kinds of neural networks [5], [6] as the matching algorithm. Frequently, the individual works consider data pre-processing techniques, such as augmenting the radio map's data representation [3], reducing the number of APs by either removing the redundant ones [7], applying radio map compression [8], [9], or reducing the number of considered samples in the database by e.g., clustering [3], [10].…”

Section: Related Workmentioning

confidence: 99%

Data Cleansing for Indoor Positioning Wi-Fi Fingerprinting Datasets

Quezada-Gaibor¹,

Klus²,

Torres-Sospedra³

et al. 2022

Preprint

View full text Add to dashboard Cite

Wearable and IoT devices requiring positioning and localisation services grow in number exponentially every year. This rapid growth also produces millions of data entries that need to be pre-processed prior to being used in any indoor positioning system to ensure the data quality and provide a high Quality of Service (QoS) to the end-user. In this paper, we offer a novel and straightforward data cleansing algorithm for WLAN fingerprinting radio maps. This algorithm is based on the correlation among fingerprints using the Received Signal Strength (RSS) values and the Access Points (APs)'s identifier. We use those to compute the correlation among all samples in the dataset and remove fingerprints with low level of correlation from the dataset. We evaluated the proposed method on 14 independent publicly-available datasets. As a result, an average of 14% of fingerprints were removed from the datasets. The 2D positioning error was reduced by 2.7% and 3D positioning error by 5.3% with a slight increase in the floor hit rate by 1.2% on average. Consequently, the average speed of position prediction was also increased by 14%.

show abstract

Section: Related Workmentioning

confidence: 99%

Data Cleansing for Indoor Positioning Wi-Fi Fingerprinting Datasets

Quezada-Gaibor¹,

Klus²,

Torres-Sospedra³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Across the existing literature, substantial effort has been made to develop efficient models and algorithms that enable efficient and accurate radio-based positioning, considering either k-NN [14], [17], [79], [80] or NN [16], [71], [81] as the positioning models, as well as to determine the necessary parameters of the radio positioning database in terms of coverage and consistency [82]. Both k-NN and NN models require a pre-existing dataset of measurements to enable localization in the given scenario, while its quality determines the lower bound for the error of the positioning system -the model can perform only as well as the data it was trained on.…”

Section: Dimensionality Reduction Techniques For Effortless Indoor Po...mentioning

confidence: 99%

From Compression of Wearable-based Data to Effortless Indoor Positoning

Klus

View full text Add to dashboard Cite

This thesis has been completed in a joint/double Doctoral Degree program between TAU and UJI as a part of the A-WEAR project for which I gratefully acknowledge the funding we have received from the European Union's Horizon 2020 Research and Innovation programme under the Marie Skłodowska Curie grant agreement No. 813278, A-WEAR. Without this support, this work would have never been possible.Furthermore, I would like to express my deepest gratitude to everyone, who has helped me to get to this point, supported me, and made me grow, both professionally and personally.First and foremost, I would like to express my sincere gratitude to my thesis supervisors. Thank you, Prof. Jari Nurmi, for letting me learn from the best, for your unlimited support, and encouragement throughout my ventures as a doctoral student. Thank you, Prof. Elena Simona Lohan, for the academic guidance, for teaching me how to grow as a researcher, as well as for all the care and encouragement you gave me throughout the years. Thank you, Dr. Carlos Granell Canut, xv for everything you did for me, your leadership, and your support. I would also like to extend my gratitude to Dr. Joaquín Torres-Sospedra (Ximo) for acting as my unofficial supervisor, for always supporting me in realizing my ideas, giving me invaluable feedback on my work, and helping me to see the paths ahead. Your guidance is highly appreciated. Furthermore, I would like to acknowledge and thank Dr. Jukka Talvitie for his support, fruitful discussions, and for letting me know about the existence of the A-WEAR project. Without you, I would have never been here. I would like to extend my thanks to Prof. Mikko Valkama, who provided me with motivation and guidance, and to my friend Asst. Prof. Bo Tan for continuous support, encouragement, and inspiration.I would like to extend my recognition to my thesis pre-examiners Prof. Fernando Javier Álvarez Franco and Dr. Christos Laoudias, as well as to my opponents Dr.Tobias Feigl, and Dr. Christos Laoudias, for investing time, energy, and effort into evaluating my thesis and offering me invaluable feedback. I am honored that scientists of such magnitude read my work.I would like to thank my colleagues and co-authors, with whom I had the pleasure to collaborate and learn from over the years. I would like to extend my thanks to all the staff of A-WEAR, TUNI, and UJI, who enabled me to successfully progress through my doctoral studies. I thank all other ESRs for creating a friendly and productive environment across the universities and for always supporting each other.Especially, I would like to thank Darwin for the fruitful collaboration that grew into a friendship. I would also like to extend my gratitude to Alex who has helped me, guided me, and always spoken his mind. To Olga and Nadia, for mutual support and for the memorable time spent together, to my friend Tomáš, as well as to Salwa, Asad, Pável, Viktoriia, and many others.Furthermore, I would like to acknowledge my colleagues and friends, with whom I have spent countless good moments he...

show abstract