雷达信号频率估计的改进R-Q算法

Improved R⁃Q algorithm for radar signal frequency estimation

在信号频率估计中,针对传统Quinn算法及Rife算法存在频率估计值点靠近量化频率点估计误差较大,Rife算法和Quinn算法(R⁃Q算法)频率估计性能不稳定的问题,提出一种改进的R⁃Q算法。该算法首先利用Quinn求出频率偏差,并基于该频率偏差得到频率估计值并进行频移;然后对频率估计值峰值点及频移0.5处的点进行离散傅里叶变换(DFT),并利用Rife算法进行幅度插值得到最终频率估计值。分析表明,该算法使待估计频率始终位于两点DFT的中心位置,有效克服了传统算法在频率估计值点靠近量化频率点估计误差较大的问题,从而提高了频率估计的准确度。仿真结果表明,该算法的性能优于R⁃Q算法和I⁃Rife算法,提高了信号频率估计的精确度,稳定性较高,计算量小,且均方根误差更接近克拉美⁃罗限(CRLB)。

In the traditional Quinn algorithm and Rife algorithm,there is a great deviation when the frequency estimation value point is close to the quantization frequency point,and the frequency estimation performance of the Rife algorithm and the Quinn algorithm(R⁃Q algorithm)is unstable in the estimation of signal frequency,so an improved R⁃Q algorithm is proposed.In this algorithm,the Quinn algorithm is used to get the frequency deviation,and the frequency estimation is obtained based on the frequency deviation,and the frequency shift is carried out.And then,the discrete Fourier transform(DFT)is applied to the peak point of the frequency estimation value and the point at the frequency shift of 0.5,and the Rife algorithm is used to interpolate the amplitude to get the final frequency estimation.The analysis shows that the algorithm makes the frequency to be estimated always located in the center of two⁃point DFT,which effectively overcomes the problem that the traditional algorithm has a large estimation error when the frequency estimation value point is close to the quantization frequency point,so as to improve the accuracy of frequency estimation.The simulation results indicate that the algorithm is better than R⁃Q algorithm and I⁃Rife algorithm.It can improve the accuracy of signal frequency estimation.It has high stability and needs lower computational complexity,and its root⁃mean⁃square error(RMSE)approaches to the Cramer⁃Rao Lower Bound(CRLB).