基于DCT-CS的脉冲星周期超快速估计方法

受快速傅里叶变换（fast Fourier transform，FFT）的影响，基于FFT和压缩感知（compressive sensing, CS）的脉冲星周期快速估计算法的计算量大。为进一步减小计算量并提高计算精度，利用离散余弦变换（discrete cosine transform，DCT)取代FFT，提出了一种基于DCT-CS的脉冲星周期超快速估计算法。在该方法中，利用DCT提取脉冲星信号的低频部分构建低频DCT矩阵；构建畸变轮廓字典并获取累积轮廓；提出了利用最大值超分辨率稀疏恢复估计脉冲星周期的方法。仿真结果表明，DCT-CS的脉冲星周期估计精度达到了3.82×10-12 s，计算时间达到了9.31 ms。与FFT-CS相比，周期估计精度提高了约16%，计算时间缩短了约37.5%，实现了实时高精度的脉冲星周期估计。

DCT-CS-based ultra fast estimation method for pulsar period

Affected by the influence of fast Fourier transform (FFT), the fast pulsar period estimation algorithm based on FFT and compressive sensing (CS) is computationally intensive. In order to further reduce the computational load and improve the calculation accuracy, a DCT-CS-based pulsar period ultra fast estimation algorithm by using discrete cosine transform (DCT) instead of FFT is proposed. In this method, the low-frequency DCT matrix is constructed by extracting the low-frequency part of the pulsar signal using DCT. A dictionary of aberration contours is constructed and the cumulative contours are obtained. A method is proposed to estimate the pulsar period using the maximum super-resolution sparse recovery. The simulation results show that the pulsar period estimation accuracy of DCT-CS reaches 3.82×10-12 s, and the calculation time reaches 9.31 ms, improve by approximately 16% and 37.5% respectively, compared with that of FFT-CS, realizing real-time and high-precision pulsar period estimation.