Decorrelating CIR fingerprint data with PCA whitening or scalar quantization before MNN training reduces indoor localization misclassification in simulated offices.
Introduction to vector quantization and its application for numerics,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.SP 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Decorrelation Deep Learning for Fingerprint-based Indoor Localization
Decorrelating CIR fingerprint data with PCA whitening or scalar quantization before MNN training reduces indoor localization misclassification in simulated offices.