Vector Tracking Algorithm Based on Adaptive Cubature Kalman Filter

基于自适应容积卡尔曼滤波的矢量跟踪算法

Received: 5 January 2018

Abstract

In the vector tracking loop, there is a great error in the output of discriminator owing to the disturbance of noise. Cubature Kalman filter is proposed to replace the discriminator to process I/Q data and generate code phase error and the carrier frequency error in this paper. The present algorithm not only can avoid the nonlinear problem of discriminator, but also can reduce the bad effect of noise. Moreover, using cubature Kalman filter to deal with the nonlinear I/Q data is beneficial to preserve the accuracy of data processing. Because noise is unknown or time-varying, the filter should have the ability to respond to the changes of environmental noise. The innovation of measurements is used to estimate the covariance matrix of measurement noise in real time. Finally, a comparison is carried out between the present algorithm and the vector tracking algorithm based on discriminator. The test results show that the code phase error and the carrier frequency error are smaller, and the accuracy of navigation solution is also higher.

摘要

当前矢量跟踪环路中，鉴别器的输出受噪声影响存在较大误差，针对该问题，提出使用容积卡尔曼滤波器代替鉴别器的算法，对I/Q支路数据进行滤波处理，输出码相位误差和载波频率误差。该算法不仅可以规避鉴别器的非线性问题，而且可以降低噪声的影响。同时，使用容积卡尔曼滤波算法处理非线性的I/Q支路数据，有效地保证了数据处理的精度。针对噪声是未知或时变的特点，采用新息协方差对量测噪声的协方差矩阵进行实时估计，提高了算法应对环境噪声变化的鲁棒性。将新算法与基于鉴别器方式的矢量跟踪算法进行对比验证，实验数据表明，改进后算法输出的码相位误差和载波频率误差更小，用户位置和速度的解算精度也更高。