基于MFOA-GRNN模型的三维定位研究

THREE-DIMENSIONAL POSITIONING BASED ON MFOA-GRNN MODEL

由于传统室内定位模型容易受到外界因素影响导致定位精度大幅下降,提出一种将多种群果蝇优化算法(Multi-population Fruit Fly Optimization Algorithm,MFOA)与广义回归神经网络(Generalized Regression Neural Network,GRNN)相结合的MFOA-GRNN三维室内定位模型。基于射频识别技术并引入多种群的果蝇优化算法用以选择GRNN的平滑参数,并通过MFOA-GRNN模型将阅读器接收信号强度与目标坐标进行对应进而判断目标位置。仿真结果表明,该模型克服了传统室内定位模型受主观因素影响大、学习效率低等不足,同时使算法的全局寻优能力得到了加强,定位精度显著提高。

Traditional indoor positioning model is vulnerable to external factors,which leads to the great decline in positioning accuracy.Therefore,this study proposes a three-dimensional indoor positioning model based on MFOA-GRNN,which combines multi-population fruit fly optimization algorithms(MFOA)and generalized regression neural network(GRNN).It was based on RPID and introduced multi-population fruit fly optimization algorithm to select the smooth parameters of GRNN.In addition,through MFOA-GRNN model,the signal strength received by the reader was corresponding to the target coordinate,so as to determine the target position.The simulation experiments show that this model overcomes the shortcomings of traditional indoor positioning model,which is influenced by subjective factors and low learning efficiency.The global optimization ability of the algorithm has been strengthened,and the positioning accuracy has been significantly improved.[HJ]