@Inbook{Alexandrou2020,
author="Alexandrou, Rafael
and Papadopoulos, Harris
and Konstantinidis, Andreas",
editor="Yang, Xin-She
and Zhao, Yu-Xin",
title="Smartphone Indoor Localization Using Bio-inspired Modeling",
bookTitle="Nature-Inspired Computation in Navigation and Routing Problems: Algorithms, Methods and Applications",
year="2020",
publisher="Springer Singapore",
address="Singapore",
pages="149--167",
abstract="Indoor localization systems (ILS) evolved over the last years, mainly due to the fact that Global Positioning Systems (GPS) lack precision or fail entirely to localize smartphone users in indoor environments. Many studies attempted to alleviate this issue by utilizing techniques developed on top of technologies, such as Bluetooth beacons and RFID that require costly installation of specialized equipment, or techniques such as Google Wi-Fi/Cell DB and fingerprinting that utilize the received signal strength intensity (RSSI) of the already existing Wi-Fi access points (APs) of a building or cellular towers of the surrounding telecommunication infrastructure. The latter approaches, however, face privacy and/or performance issues since they gain knowledge regarding the requested user location during the indoor localization process and/or require high energy, storage and computational power resources in order to localize the user in situ the smartphone device. In this chapter, we introduce a bio-inspired modeling approach for addressing all the aforementioned concerns. In particular, an artificial neural network is used to model an area's RSSI data at the server side, which is then forwarded to the client side to predict the user's current location on the smartphone device. The proposed approach improves the smartphone performance during user navigation while maintaining a fine-grain localization accuracy, but also preserving the user's privacy with respect to conventional fingerprinting approaches. Our experimental studies demonstrate the superiority of the proposed approach on real datasets of Wi-Fi traces.",
isbn="978-981-15-1842-3",
doi="10.1007/978-981-15-1842-3_7",
url="https://doi.org/10.1007/978-981-15-1842-3_7"
}