双自适应BP算法在智能手环中的应用

Application of Dual-adaptive BP Algorithm in Intelligent Bracelet

智能手环体征信息检测的准确性与及时性直接影响到它的使用和推广。通过多传感器数据融合技术对人体体征信息进行检测是主流趋势。针对标准BP神经网络算法存在的收敛速度慢、容易陷入局部极值的缺陷,增加了自适应动量法和自适应学习率调整法以改进传统BP算法,给出了改进后的BP网络模型及参数设计,并应用于智能手环体征检测工作中。仿真实验证明,改进的算法能够使BP网络快速收敛,且提高了手环系统报警的准确率,减少了误判的发生。

The accuracy and timeliness of smart bracelet sign detection directly affect its use and promotion.For human body sign information detection,the multi-sensor data fusion technology is needed.In view of the defects of slow convergence rate and easy to fall into local extremum,the adaptive momentum method and adaptive learning rate adjustment method are added to improve the traditional BP algorithm.The improved BP network model and parameter design are given,and it is applied to the detection of intelligent hand loop（smart BP）.The simulation experiment results show that the performance of the improved algorithm is obviously superior to the traditional BP algorithm.It can make the BP network converge quickly,and improve the accuracy of the alarm system of the hand loop system and reduce the occurrence of miscarriage of justice.