Indoor Visible Light Positioning Method Based on Extreme Learning Machine Neural Network

Table 1. RMSE of models with different number of hidden layer neuronsunit: cm
View table
Table 1. RMSE of models with different number of hidden layer neuronsunit: cm
Number of neurons RMSE
240 2.36
245 2.47
250 2.13
253 1.75
255 1.36
260 1.17
261 1.30
263 1.48
265 1.65
270 1.42

Table 2. Simulation parameters
View table
Table 2. Simulation parameters
Parameter Value
$P_{t}$ /W 10
$ψ_{c}$ /（°） 90
$T_{s}$ 1
$g$ 10
$A$ /cm² 1
$ϕ_{1 / 2}$ /（°） 30

Table 3. Positioning errors of different methodsunit: cm
View table
Table 3. Positioning errors of different methodsunit: cm
method
Positioning
Max positioning error Average positioning error
ELM 6.44 1.17
SVM 16.89 3.74
BP 63.60 21.23
GA-BP 10.29 2.72

Table 4. Positioning time of different methodsunit: s
View table
Table 4. Positioning time of different methodsunit: s
Positioning method Training time of fingerprint data Average positioning time
ELM 0.0687 0.03594
SVM 0.1094 0.09375
BP 44.0703 0.09063
GA-BP 74.9375 0.09562

Tools

Get Citation

Copy Citation Text

Ling Qin, Dongxing Wang, Fengying Wang, Xiaoli Hu. Indoor Visible Light Positioning Method Based on Extreme Learning Machine Neural Network[J]. Laser & Optoelectronics Progress, 2022, 59(3): 0304002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites