Heterogeneous Transfer Learning for Wi-Fi Indoor Positioning Based Hybrid Feature Selection

Abstract

This paper presents the application of heterogeneous transfer learning (HetTL) methods which consider hybrid feature selection to reduce the training calibration effort and the noise generated by fingerprint duplicates obtained from multiple Wi-Fi access points. The Cramer–Rao Lower Bound analysis (CRLB) was also applied to evaluate and estimate a lower limit for the variance of a parameter estimator used to analyze positioning performance. We developed two novel algorithms for feature selection in fingerprint-based indoor positioning problems (IPP) to enhance positioning performance in the target domain with the HetTL. The algorithms comprised two scenarios: (i) a principal component analysis-based approach (PCA-based) and (ii) a hybrid approach that takes both PCA and correlation effect analysis into account (hybrid scenario). Accordingly, a new feature vector was constructed by retaining only the most significant predictors, and the most efficient feature dimensions were also determined by using a hybrid-based approach. Experimental results showed that the hybrid-based proposed algorithm has the minimum mean absolute error. The CRLB analysis also showed that the number of Wi-Fi access points could affect the lower bound location estimation error; however, identifying the most significant predictors is an effective approach to improve positioning performance.

Feature Selection and Model Evaluation for Threat Detection in Smart Grids

“…For example, in scenarios with limited data, limited computational resources, or high-dimensional input data, feature selection techniques can still contribute to a reduction in computational complexity and an improvement in model performance [52]. Additionally, in transfer learning scenarios, where pre-trained deep learning models are tuned for specific tasks, feature selection can be used to more efficiently adapt the learned representations to the target task [53,54].…”

Data Fusion Methods for Indoor Positioning Systems Based on Channel State Information Fingerprinting

“…Furthermore, indoor positioning-based Wi-Fi RSS fingerprints have been characterized by low-dimensional feature spaces and a poor spatial resolution, which directly degenerates the indoor positioning performance [70]. In summary, the system fails to achieve the desired accurate and robust positioning estimates due to four critical predictors associated with the RSS-based fingerprint that determines the quality of the IPS: (i) high temporal signal fluctuations, (ii) RSS measurements highly susceptible to the effect of a typical indoor environment, (iii) low-dimensional feature spaces, and (iv) requirements for a large size of labeled samples [37], which is both costly and labor-time-intensive [37][38][39][40][41]. Moreover, RSS is also highly dependent on the used Wi-Fi chipset and how it estimates and reports the RSS value.…”

“…While principal component analysis is important for minimizing computational complexity, removing significant features has a critical impact on positioning accuracy and negatively impacts positioning performance. A "base model," according to [38], consists of only two principal components to represent the variational distributions of the target's prediction, accounting for approximately 16% of the total variance explainability of the model, though [38] achieved 56%. Even though the base model has a lower variance explainability ratio of 16% versus 56%, this finding is consistent with the findings in [38] that lower feature space dimensions can improve computational cost and model simplicity.…”

“…This experimental result [37] confirms that the temporal variations in signal distributions have a significant impact on indoor positioning performance-based RSS fingerprints, owing to the effects of multipath, NLOS, and channel conditions such as fading, shadowing, and scattering. Indoor positioning-based RSS fingerprints (FBIP-RSS), on the other hand, have been characterized as having poor spatial resolution and low dimensional feature spaces, which directly degrades indoor positioning accuracy [38]. As a result, the system fails to achieve the desired accurate and robust positioning estimates due to four critical predictors associated with the RSS-based fingerprint that determine the quality of the IPS: (i) high temporal signal fluctuations, (ii) RSS measurements highly susceptible to the effect of a typical indoor environment, (iii) low dimensional feature spaces, and (iv) requirements for large size labeled samples [37][38][39][40][41], which is both costly and labor-intensive.…”

Co-Occurrence Fingerprint Data-Based Heterogeneous Transfer Learning Framework for Indoor Positioning

“…In [ 52 ], a common feature space was constructed by resorting to a cross-domain mapping, allowing to adopt domain adaptation methods for further knowledge transfer. Furthermore, [ 53 ] proposed a new feature extraction scheme by retaining only the most significant predictors; they selected the most efficient feature dimensions by utilizing a hybrid-based approach to reduce the training calibration efforts. However, the aforementioned works neglected a long-lasting problem on how to determine whether (and how) the source domain knowledge is valuable for the target positioning task.…”