基于卡尔曼滤波器的IEEE 1588时钟同步算法

时钟同步是网络化分布式测试与控制系统中的一项重要指标。在基于IEEE 1588协议的主从时钟同步中,时钟偏差和时钟漂移的精确测量是主从时钟同步的重要保证。提出了基于二阶卡尔曼滤波器加速运动模型的时钟同步算法,该算法以同步消息包中的时间戳来获取观测值,通过卡尔曼滤波器算法对主从时钟之间的时钟偏差、时钟漂移以及时钟漂移变化率进行估计,使用估计值对从时钟进行补偿与修正。该算法能够消除从时钟的不稳定性对时钟同步的影响。实验结果表明,在时钟同步中引入卡尔曼滤波算法能够显著提高时钟同步精度。

IEEE 1588 clock synchronization algorithm based on Kalman filter

Clock synchronization is a fundamental issue in network distribution test and control system.In IEEE 1588 precision time protocol(PTP) based on clock synchronization,the clock offset and clock skew between master clock and slave clock are critical parameters in master–slave structured network.In this paper,a method of clock synchroniza-tion based on 2 orders Kalman filtering accelerated motion model is proposed.In the proposed algorithm,the timestamp in the exchanged massages is used to estimate the clock offset,clock skew and clock aging rate between the slave clock and the master clock.The estimated values are used to synchronize the slave clock and modify the slave clock model.The proposed algorithm can eliminate the impact on the instability of slave clock.The experimental results demonstrate that the proposed algorithm can significantly improve the accuracy of clock synchronization.