Lightweight Workload Fingerprinting Localization Using Affinity Propagation Clustering and Gaussian Process Regression

Abstract

Fingerprinting localization approach is widely used in indoor positioning applications owing to its high reliability. However, the learning procedure of radio signals in fingerprinting is time-consuming and labor-intensive. In this paper, an affinity propagation clustering (APC)-based fingerprinting localization system with Gaussian process regression (GPR) is presented for a practical positioning system with the reduced offline workload and low online computation cost. The proposed system collects sparse received signal strength (RSS) data from the deployed Bluetooth low energy beacons and trains them with the Gaussian process model. As the signal estimation component, GPR predicts not only the mean RSS but also the variance, which indicates the uncertainty of the estimation. The predicted RSS and variance can be employed for probabilistic-based fingerprinting localization. As the clustering component, the APC minimizes the searching space of reference points on the testbed. Consequently, it also helps to reduce the localization estimation error and the computational cost of the positioning system. The proposed method is evaluated through real field deployments. Experimental results show that the proposed method can reduce the offline workload and increase localization accuracy with less computational cost. This method outperforms the existing methods owing to RSS prediction using GPR and RSS clustering using APC.

A Survey of Smartphone-Based Indoor Positioning System Using RF-Based Wireless Technologies

“…Some of the examples of probability-based fingerprinting localization using Wi-Fi as a signal source are illustrated in [ 123 , 142 , 149 , 150 ]. Similarly, the literature that employ BLE for probability-based fingerprinting are presented in [ 59 , 151 ].…”

“…This work estimated the hyperparameters by using the subspace trust-region method and shows that location estimation with a radio map built using GPR is better than that of Horus fingerprinting method [ 153 ]. The GPR-based IPS in [ 151 ] utilizes BLE beacons for localization where the Hlhyperparameters are optimized employing limited memory BFGS-B [ 154 ]. Here, the predicted RSS data is further preprocessed for RSS clustering, where the final localization result is obtained with the minimized offline workload and reduced online computational complexity.…”

An Indoor Positioning and Navigation System Using Named Data Networking

“…Their optimized CNN replaces general matrix multiplication, which reduces computational complexity. Subedi et al [31] proposed an affinity propagation clustering (APC)-based fingerprinting localization system with Gaus-sian process regression to estimate RSS values in the offline phase and used APC to reduce the search space during the online phase; this method improved the accuracy and reduced the computational load.…”

Improving Indoor Fingerprinting Positioning With Affinity Propagation Clustering and Weighted Centroid Fingerprint

“…In the conventional fingerprinting or flat-based fingerprinting localization, the computational complexity grows with the size of the radio map (number of RPs on the radio map) [3], [22]. For N RPs on the testbed, the computational complexity of traditional fingerprinting localization (linear matching process) is O(N ) [40], [41].…”