基于波形捷变的多传感器机动目标跟踪

针对现代战场中目标往往采用机动方式运动的情况,为了提高目标跟踪的准确性和精确性,结合多传感器数据融合的优点,提出了一种基于波形捷变的多传感器机动目标跟踪方法。该算法通过波形捷变来改变量测的精度。首先在现有文献的基础上,将波形捷变方式推广到二维空间,把雷达量测的克拉美罗下限（Cramer-Rao lower bound,CRLB）近似为量测误差协方差,由于该CRLB是关于发射波形参量的,从而把雷达跟踪的信号处理与数据处理结合在一起,通过波形参量的动态选择得到量测误差协方差的最小值。从而在整个雷达跟踪过程中提高信噪比（signal to noise ratio,SNR）,降低量测误差。其次,在数据处理上,采用多传感器数据融合及粒子滤波进一步提高机动目标跟踪的精度。最后,将该算法与传统的Kalman滤波、粒子滤波及只对一维空间的量测采用波形捷变的算法和交互多模型方法（interacting multiple model,IMM）进行仿真比较,仿真结果显示该算法对机动目标的跟踪精度显著提高。

Maneuvering target tracking based on waveform agility with multi-sensor

To deal with the case that modern battlefield often has maneuvering targets, a method of maneuvering target tracking is proposed based on waveform agility with multi-sensors to improve the veracity and accuracy of the performance. First, we establish the Cramer-Rao lower bound （CRLB） function of the measure errors of the sensors. Since the function consists of the parameters of the transmitted waveform, which can be selected adaptively, we can minimize the covariance of the measurements. Then the tracking precision is improved and the signal-to-noise ratio （SNR） is increased. The algorithm given above takes the measurements from two-dimensions space to realize target tracking. And we compare it with the conventional Kalman filtering, particle filtering and the method with waveform-agility which only has measurements from the one dimension space. Simulation results show that the proposed algorithm provides better tracking performance.