利用宽带信号统计的鱼群资源量反演方法

Fish abundance inversion method using broadband signal statistics

针对鱼群资源量反演问题,建立了一种基于宽带回波信号统计的反演方法。首先考虑自然状态下鱼群活动的随机性,并通过回波信号波形简化,实现了对鱼群宽带回波脉冲压缩输出信号的建模。然后以信号强度变异系数作为统计特征量,完成了统计特征量与鱼群资源量的数值关系模型的推导。仿真与海试实验结果表明该方法能有效实现鱼群资源量的反演,且比已有的窄带反演方法误差更小、统计所需样本更少。与主流的回波积分法相比,该方法不依赖鱼群先验知识,且无需计算声呐信号的绝对声学量,具有很好的应用潜力。

To the problem of fish abundance inversion,a method based on broadband echo signal statistics is established.Firstly,considering the randomness of fish shoal activity in nature state,and by simplifying the echo signal waveform,the modeling of pulse-compression output signal of fish broadband echoes is realized.Then,by using the coefficients of variation of signal strength as the statistical feature quantity,the numerical relationship between the statistical feature and the fish abundance is derived.The simulation and sea experiment results show that the proposed method can effectively achieve fish abundance inversion,and it’s more precise and requires fewer samples when compared with the existing narrowband inversion method.Compared with the mainstream echo-integration method of fisheries acoustics,this method does not rely on prior knowledge of fish school,and does not need to calculate the absolute acoustic quantity of the sonar signals,which shows great application prospects.