Title :
Applied Research on Discriminative-Adaptive Neural Network Algorithm in Indoor Positioning System
Author :
Du, Youfu ; Zhao, Ming ; Hu, Yanghong
Author_Institution :
Sch. of Comput. Sci., Yangtze Univ., Jingzhou, China
Abstract :
In this paper, a three-layer Discriminative-Adaptive Neural Network algorithm (DANN) is proposed in indoor Positioning System. By using Multivariate Discriminant Analysis, a weight matrix could be acquired, then it could input these weighting signals into the Neural Network for training, and it could get a minimum mean square error, finally, we could get the Minimum error positioning results. Experimental results show that using DANN algorithm can obviously improve the indoor positioning precision and speed.
Keywords :
Global Positioning System; least mean squares methods; matrix algebra; neural nets; telecommunication computing; MMSE; indoor positioning system; minimum mean square error; multivariate discriminant analysis; three-layer discriminative-adaptive neural network algorithm; weight matrix; weighting signals; Algorithm design and analysis; Educational institutions; Fingerprint recognition; Instruments; Radar tracking; Training; Wireless LAN; Discriminant Analysis; neural networkt; positioning system; weight matrix;
Conference_Titel :
Genetic and Evolutionary Computing (ICGEC), 2011 Fifth International Conference on
Conference_Location :
Xiamen
Print_ISBN :
978-1-4577-0817-6
Electronic_ISBN :
978-0-7695-4449-6
DOI :
10.1109/ICGEC.2011.83
