高阶运动目标的长时间相参积累算法

Long time coherent integration algorithm for moving targets with high order motion

针对具有任意阶运动的目标的长时间相参积累问题,提出一种基于多维非均匀快速傅里叶变换(non-uniform fast Fourier transform,NUFFT)的长时间相参积累算法。该算法先在快时间频域沿慢时间维利用多维NUFFT实现运动补偿,然后通过快速傅里叶逆变换(inverse fast Fourier transform,IFFT)最终实现相参积累。该算法积累性能接近理论最优且计算量小于已有算法。特别地,对于具有加加速度的运动目标进一步提出基于Wigner-NUFFT的相参积累算法,该算法相比多维NUFFT,计算量大大减小,但对积累前单个脉冲的信噪比提出更高要求。仿真结果证明了所提算法的有效性。

A long time coherent integration algorithm based on multi-dimensional non-uniform fast Fourier transform （NUFFT） is proposed for moving targets with arbitrary orders of motion. The presented algorithm firstly utilizes multi-dimensional NUFFT to realize motion compensation in the fast-time frequency domain, and then accomplishes coherent integration via inverse fast Fourier transform （IFFT）. The performance of this algorithm is almost optimal while the computational complexity is less than that of the existed algorithms. Especially, another new algorithm with Wigner-NUFFT is used to deal with a target with a jerk. Compared to the algo-rithm with multi dimensional NUFFT, the algorithm with Wigner-NUFFT has much lower computational corn-plexity. However, the disadvantage is that a higher signal-to noise ratio is needed. The simulation results demonstrate the effectiveness of the proposed algorithm.