Abstract
Fingerprint-based positioning techniques are a hot research topic because of their satisfactory accuracy in complex environments. In this study, we adopted the deep-learning-based long-time-evolution (LTE) signal fingerprint positioning method for outdoor environment positioning. Inspired by state-of-the-art image classification methods, a novel hybrid location gray-scale image utilizing LTE signal fingerprints is proposed in this paper. In order to deal with signal fluctuations, several data enhancement methods are adopted. A hierarchical architecture is put forward during the deep neural network (DNN) training. First, the proposed positioning technique is pre-trained by a modified Deep Residual Network (Resnet) coarse localizer which is capable of learning reliable features from a set of unstable LTE signals. Then, to alleviate the tremendous collection workload, as well as further improve the positioning accuracy, by using a multilayer perceptron (MLP), a transfer learning-based fine localizer is introduced for fine-tuning the coarse localizer. The experimental data was collected from realistic scenes to meet the requirement of actual environments. The experimental results show that the proposed system leads to a considerable positioning accuracy in a variety of outdoor environments.
Highlights
In recent years, smartphone-based positioning has been attracting attention due to the increasing number of equipped sensors and rapid development of various positioning techniques
The basic idea of the signal fingerprint-based technique is to find the geo-tag signal features, such as channel state information (CSI), received signal strength indication (RSSI), reference signal receiving power (RSRP), and reference signal receiving quality (RSRQ), and match them against a pre-defined signal database to find the location of user equipment (UE) [10,11,12]
The way of collecting the dataset had an impact on positioning, and in order to study the impact of mobility smartphone with an Android system, which was equipped with a ubiquitous built-in chip that could receive the LTE signal
Summary
Resnet is is well-trained, well-trained, transfer transfer learning learning is is leveraged leveraged to to fine-tune fine-tune the the neural neural network. When training training the the transfer transfer learning, learning, the the parameter parameter of Resnet remains remains unchanged. Another consisting ofofseveral several hidden layers is leveraged to further improve the positioning consisting hidden layers is leveraged to further improve the positioning accuracy.accuracy. Its feedIts feed-forward and backward propagation are the same as the aforementioned layer. Forward and backward propagation are the same as the aforementioned MLP layer. The Resnet and and use the same training set and test set for the purpose of comparing the coarse localizer use the same training set and test set for the purpose of comparing the coarse localizer and the fine and the fine localizer accuracy. The proposed proposed transfer transfer learning learning fine fine localizer localizer structure
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
