Abstract
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 a large number of training data points to locate accurately, we have realized high-precision positioning for 100 test points with only 20 training points in a 1.8 m × 1.8 m × 2.1 m localization area. In order to verify the adaptability of the MMBP algorithm, we experimentally demonstrate two different training data acquisition methods adopting either even or arbitrary training sets. In addition, we also demonstrate the positioning accuracy of the traditional RSS algorithm. Experimental results show that the average localization accuracy optimized by our proposed algorithm is only 1.88 cm for the arbitrary set and 1.99 cm for the even set, while the average positioning error of the traditional RSS algorithm reaches 14.34 cm. Comparison indicates that the positioning accuracy of our proposed algorithm is 7.6 times higher. Results also show that the performance of our system is higher than some previous reports based on RSS and RSS fingerprint databases using complex machine learning algorithms trained by a large amount of training points.
Highlights
In recent years, with the rapid development of wireless sensor networks and the Internet of Things, indoor positioning technology has been used in many fields, such as shopping guidance, intelligent robots, and so on
Traditional positioning techniques based on radio wave communication, such as the extensively used global positioning system (GPS), suffer from multipath fading and severe attenuation, which lead to large positioning errors in the indoor environment
As we mainly focus on the question of how to achieve high-accuracy positioning with sparse training points, number of hidden layers is a critical factor determining the complexity of neural networks, and complex neural networks need a lot of training data to train
Summary
With the rapid development of wireless sensor networks and the Internet of Things, indoor positioning technology has been used in many fields, such as shopping guidance, intelligent robots, and so on. In [12], the average positioning error was only 6.39 cm in the simulation stage with the Levenberg-Marquardt algorithm and the number of training data was 2630 within a 5 m × 5 m × 3 m area Earlier this year, Hsu et al realized 3.65 cm accuracy in a 1 m × 0.9 m triangular unit cell [13]. C proposed a lightweight fingerprint-based indoor positioning algorithm using generalized regression neural network (GRNN) to approach the time-consuming and labor-intensive process of fingerprint surveying They selected 140 training points in the area of 3 m × 5 m and the average positioning error was 8.7 cm. In [16], Guo X proposed a two-layer fusion network (TLFN) indoor localization method using the same visible light positioning scenario as [15] and the average positioning error was 5 cm. Compared with the results based on RSS fingerprint in [15], our results are more accurate and the size of the training set is only 1/11 of that reported in Reference [11]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
