Combining with characteristic of low cost and high precision of radio frequency identification（ RFID）technology

an artificial neural network method based on particle swarm optimization（ PSO） is proposed. The method uses PSO to optimize weight and threshold of artificial neural network（ ANN）

which avoids shortcoming of traditional ANN prediction results easy to fall into local optimum. Relationship between received signal strength indication value and position coordinate of tag signal is established. At the same time

in order to reduce interference of the environmental factors on signal

the Gaussian filtering is adopted to process the RSSI value

so as to increase positioning precision. The experimental result demonstrates that the proposed method has better performance. Average localization error of the method is reduced by 11. 72 cm

compared with the traditional ANN

and stability of the algorithm is better.