High precision frequency estimation algorithm for autocorrelation weighted fusion

In order to ensure the performance of radar countermeasure reconnaissance system in complex electromagnetic environment, a multi-segment signal frequency estimation algorithm based on autocorrelation weighted fusion was proposed. Autocorrelation processing was performed on each segment of the noisy signal to obtain a high SNR sinusoidal signal with zero initial phase and the same frequency as the original signal. The arcsine operator was used to construct a support matrix to perform real-time weighted fusion of the autocorrelated signals. The reference signal was established on the basis of the rough estimation, and the high-precision frequency estimation result was obtained by minimizing the error function. The simulation results show that, compared to existing algorithms, the algorithm not only improves the accuracy significantly, but also has stable estimation performance under the conditions of different SNRs, signal lengths and signal anomalies, while satisfying the low computational cost, which provides a reference for radar countermeasures intelligence reconnaissance based on multi-sensors.