Unknown Source Positioning Based On Extreme Learning Machine

Wireless location technology based on neural network has always been a hot research field at home and abroad.As the urban radio environment becomes more and more complicated,more and more interferences are caused to the traditional positioning methods,resulting in larger positioning errors.Positioning technology based on neural network often requires more sample data,training time is longer,work efficiency is lower.Therefore,it is necessary to study a wireless positioning technology of neural network with high positioning speed and high positioning accuracy.In this paper,we invent an improved localization algorithm of nuclear limit learning machine.This algorithm solves the problem that the location results always subject to external environment interference and the long time training of neural network wireless positioning algorithm.Extreme learning machine positioning methods are using non-measurement positioning methods,so extreme learning machine belongs to the feed-forward neural network which has great anti-interference ability.Firstly,the training data is obtained by the method of multiple measurements at the same location.Then,the data obtained at the same position is divided into a sample subspace and the characteristics of the sample subspace are extracted to replace the original training data.At the same time,the kernel extreme learning machine algorithm is improved by using the matrix approximation and matrix extension theory.Finally,the obtained processed training data is trained by the improved kernel extreme learning machine,and the positioning prediction model is obtained.The simulation results show that the improved kernel extreme learning machine has shorter training time and the positioning speed is faster under the same data set.In the case of the same noise interference,the algorithm has less prediction error.The simulation results show that under the same data set,the improved kernellimited learning machine has the advantages for short training time and high positioning speed.Under the same noise interference,the prediction error of this algorithm is negligible.It is verified that this algorithm can not only improve the training speed and positioning speed,but also effectively reduce the noise interference and improve the positioning accuracy.