DeepLocate: Smartphone Based Indoor Localization with a Deep Neural Network Ensemble Classifier

Abstract

A quickly growing location-based services area has led to increased demand for indoor positioning and localization. Undoubtedly, Wi-Fi fingerprint-based localization is one of the promising indoor localization techniques, yet the variation of received signal strength is a major problem for accurate localization. Magnetic field-based localization has emerged as a new player and proved a potential indoor localization technology. However, one of its major limitations is degradation in localization accuracy when various smartphones are used. The localization performance is different from various smartphones even with the same localization technique. This research leverages the use of a deep neural network-based ensemble classifier to perform indoor localization with heterogeneous devices. The chief aim is to devise an approach that can achieve a similar localization accuracy using various smartphones. Features extracted from magnetic data of Galaxy S8 are fed into neural networks (NNs) for training. The experiments are performed with Galaxy S8, LG G6, LG G7, and Galaxy A8 smartphones to investigate the impact of device dependence on localization accuracy. Results demonstrate that NNs can play a significant role in mitigating the impact of device heterogeneity and increasing indoor localization accuracy. The proposed approach is able to achieve a localization accuracy of 2.64 m at 50% on four different devices. The mean error is 2.23 m, 2.52 m, 2.59 m, and 2.78 m for Galaxy S8, LG G6, LG G7, and Galaxy A8, respectively. Experiments on a publicly available magnetic dataset of Sony Xperia M2 using the proposed approach show a mean error of 2.84 m with a standard deviation of 2.24 m, while the error at 50% is 2.33 m. Furthermore, the impact of devices on various attitudes on the localization accuracy is investigated.

Indoor Positioning System using WKNN and LSTM Combined via Ensemble Learning

“…Several research [22]- [25] used an ensemble learning approach where the outcomes of several positioning algorithms are combined into one single output. In this study, we provide a different ensemble method which combines two positioning algorithms, namely WKNN and LSTM.…”

“…A research by Li et al provide an ensemble approach to fuse Differential Time Difference Of Arrival (DTDoA) with RSS value [21]. Some research provide a way to combine the results from several positioning algorithm such as KNN [22]- [24] and neural network [25]. Research by Hayashi divide the environment into several area and each area has its own positioning algorithm.…”

A Comprehensive Analysis of Magnetic Field Based Indoor Positioning With Smartphones: Opportunities, Challenges and Practical Limitations

“…However, other research works aim to utilize deep learning models that are trained on the magnetic field data. For example, research [30], [31] leverage an ensemble of neural networks which are trained on the magnetic field data to predict a user's current position. An image-based approach is adopted in [32] where the magnetic field patterns are stored as an image to train several CNN.…”

A New Dataset of People Flow in an Industrial Site with UWB and Motion Capture Systems

“…The use of magnetic data from several smartphones to generate a magnetic pattern is used for better position detection [24] as well as a multisensor fusion, based on the magnetic field [25]. The use of deep neural networks (DN) to perform magnetic field-based indoor localization using heterogeneous devices [26] allows a more reliable system. An efficient data collection method based on walking and responds to all challenges in the magnetic field as been made for indoor localization [27].…”