CAJ | 학술논문

对于传统的对移动台的定位，提出了一种基于粒子群（ PSO ）优化神经网络的算法。这一PSO-BP 算法首先利用 PSO 对神经网络传统的目标函数及参数进行优化，再利用改进后的 BP 神经网络对非视距误差（ NLOS ）进行修正，最后利用算法 LS 进行移动台的定位。仿真结果表明，该基于 PSO的神经网络定位算法寻优效果稳定，预测误差小，具有可行性。
대우전통적대이동태적정위，제출료일충기우입자군（ PSO ）우화신경망락적산법。저일PSO-BP 산법수선이용 PSO 대신경망락전통적목표함수급삼수진행우화，재이용개진후적 BP 신경망락대비시거오차（ NLOS ）진행수정，최후이용산법 LS 진행이동태적정위。방진결과표명，해기우 PSO적신경망락정위산법심우효과은정，예측오차소，구유가행성。
A location algorithm based on PSO ( Particle Swarm Optimization ) using BP ( Back Propagation ) neural network for the traditional location of mobile station in this paper . The algorithm is using PSO to optimize the traditional objective function and pa-rameters of BP , and then correct the NLOS errors . At last , the positions of MS can be estimated by LS′s algorithm . The results in-dicate that the localization algorithm is better than others in the stability and the prediction error is smaller .