High-Precision Indoor Visible Light Positioning Using Modified Momentum Back Propagation Neural Network with Sparse Training Point

• DOI: 10.3390/s19102324
• Publication Date: 2019-05-20T15:05:07Z
• Authors: Haiqi Zhang; Jiahe Cui; Lihui Feng; Aiying Yang; Huichao Lv; Bo Lin; Heqing Huang
• Keywords: RSS; Training set; Adaptability; Data set; Backpropagation

In this letter, we propose an indoor visible light positioning technique using a Modified Momentum Back-Propagation (MMBP) algorithm based on received signal strength (RSS) with sparse training data set. Unlike other neural network algorithms that require large number of points to locate accurately, have realized high-precision for 100 test only 20 in 1.8 m × 2.1 localization area. order verify the adaptability MMBP algorithm, experimentally demonstrate two different acquisition methods adopting either even or arbitrary sets. addition, also accuracy traditional RSS algorithm. Experimental results show average optimized by our proposed is 1.88 cm set and 1.99 set, while error reaches 14.34 cm. Comparison indicates 7.6 times higher. Results performance system higher than some previous reports fingerprint databases complex machine learning trained amount points.