广义的自适应相干累积算法

根据最小均方误差(LMS)算法、自适应相干累积(ACI)算法及推广的自适应相干累积(GACI)算法在原理上的相关性和推广关系，提出了广义的自适应相干累积(IGACI)算法。它是在GACI的基础上，在算法的权系数迭代公式中引入更多的动量因子，即利用权系数的更多的历史信息，使被检测的单频或调频脉冲信号，在脉冲期间，实现自适应相干累积，从而完成低输入信噪比条件下的信号检测，而无需或很少需要知道信号的先验信息。理论分析和仿真结果证明在低信噪比的情况下，其性能要优于GACI算法。

An Improved Generalized Adaptive Coherent Integrator Algorithm

GACI (Generalized Adaptive Coherent Integrator) algorithm is quite effective in detecting pulse signals. We aim to improve it further and present in this paper improved GACI algorithm, called by us IGACI algorithm. In IGACI algorithm, momentum factors, more than those in GACI algorithm, are introduced into the iteration formulas for weight coefficients. This greater introduction of momentum factors allows the utilization of more history information available in weight coefficients during impulse period, and accordingly the sinusoidal and LFM (Linear Frequency Modulated) signals to be detected can be adaptive coherent integrated; thus, under the condition of low input SNR (signal to noise ratio), signal detection can be accomplished without requiring knowledge of a priori information or with only a little knowledge of a priori information required. Both GACI and IGACI algorithms are used for detecting the LFM signals in white Gaussian noise. Simulation results confirm preliminarily that IGACI algorithm is indeed better than GACI algorithm, and show that IGACI algorithm is feasible.