重叠平均法性能分析及帧头辅助
Performance Analyses for Overlap Averaging Method with Frame Head Assistance
-
摘要: 对重叠平均法性能进行定量分析,建立了检测概率和算法参数的数学关系,通过推导可知,当利用重叠平均法加快搜索速度时,不可避免灵敏度损失的问题,即重叠平均点数扩大一倍则检测性能恶化约3 dB. 为了适应低信噪比环境和高数据率系统,提出了帧头辅助的方法,该方法延长相关积累时间增加积累增益以补偿重叠平均法的灵敏度损失,且实现较为简单. 通过仿真对建立的检测概率和算法参数的数学关系进行了验证,显示了帧头辅助的重叠平均快速捕获算法的优点,即利用4 ms八点重叠平均几乎达到了2 ms直接二点平均的检测性能,但数据处理量减小1/2.Abstract: The quantitative performance analysis of overlap averaging method is given for the first time in this paper. As result of the analysis, the mathematic formula of detection probability was derived. From mathematic deduction, it was revealed that the sensitivity deterioration couldn't be ignored when the overlap averaging method was used to speed up the search. Specifically, the sensitivity deterioration would be about 3 dB when the number of overlap averaging points was doubled. To be able to still work in low SNR environment, some compensation measures should be taken for overlap averaging method. A new measure of taking frame head to assist acquisition was proposed. The overlap averaging method with frame head assistance could improve accumulation gain because it extended the correlative integration time simply. Monte Carlo simulation has been implemented. The results show that the accuracy of derived mathematic formula is satisfied and the performance of frame head assisted overlap averaging method is good. The detection performance of 4 ms integration using eight-point overlap averaging method is almost the same as that of 2 ms integration using two-point direct averaging method, while the amount of data processing is reduced by half.
下载: