基于同频等长信号融合的频率估计快速迭代算法

Fast iterative algorithm of frequency estimation based on information fusion among signals with same frequency and length

多段同频等长信号十分常见,对其进行信息融合能有效提高信号处理的精度,尤其适用于低信噪比、被测频率持续时间短的情况。为提高频率估计的精度、实时性和扩展已有方法的适用范围,给出了一种基于多段同频等长信号融合的快速迭代算法。在该算法中,设计相位差补偿因子矩阵以克服各段信号之间相位不连续;生成搜索频率序列以修正相位差补偿因子矩阵中的未知参量并得到具有特定形式的功率谱矩阵;采用间接迭代计算以改善算法实时性在说明上述技术措施的应用原理时,给出了5项重要性质并作为该算法正确性的数学证明,针对多种应用环境状态进行了仿真实验,结果表明,该算法具有普适性、抗噪性好、频率估计精度比现有方法有较大提高,具有重要的理论和应用价值。

Multi-section signals with the same frequency and length is common, and information fusion among them can increase precision of signal process obviously, especially under the condition of low SNR and short duration of measured frequency. To increase precision and real-time performance of frequency estimation and extend applicable range of existing methods, a fast iterative algorithm for frequency estimation of multl-section signals with the same frequency and length is proposed. In this algorithm, a phase-difference compensating factor matrix was designed to overcome phase discontinuity among different section signal. A search frequency array was generated to modify unknown parameters in the phase-difference compensating factor matrix and a power spectrum matrix with certain form was obtained. Indirect iterative computation was adopted to improve real-time performance of the algorithm. In the description of application principle of above technical measures, five important properties were proved as mathematical justification of algorithm correctness. Simulation were carried out under many kinds of complicated application environment, and the simulation result shows the better universality, noise immunity and higher frequency estimation precision of the algorithm.