利用神经网络的海杂波幅度分布参数估计方法

海杂波是制约对海雷达探测性能的主要因素之一,掌握其特性,具有十分重要的意义。经典海杂波统计模型在参数估计方法上以传统统计学理论为基础,在样本数较少的情况下,估计结果往往较差,导致建模准确度下降。此外,在复杂非均匀探测背景下,难以实现海杂波模型参数的准确实时估计。针对该问题,文章将深度神经网络模型引入海杂波参数估计领域,通过构建合理的模型,使其具备海杂波幅度分布模型的高精度参数估计能力。该方法采用直方图统计的方法进行数据预处理,合理划分输入数据标签的分组区间,构建数据集训练神经网络,并利用测试数据得到神经网络估计结果。仿真数据和X波段IPIX雷达实测数据验证结果表明,与传统数理统计估计方法相比,该算法明显提升了海杂波统计模型参数估计精度。

Amplitude Distribution Parameter Estimation Method of Sea Clutter Using Neural Network

Sea clutter is one of the main factors that restrict the detection performance of the sea radar, and mastering itscharacteristics are very important. The classical statistical model of sea clutter is based on the traditional statistical theory,in the case of a small number of samples, the estimation results tend to be poor, and resulting in a decrease in modeling ac?curacy. In addition, in the complex and non-uniform detection background, it is difficult to achieve accurate real-time esti?mation of the sea clutter parameters. Aiming at this problem, in this paper deep neural network model was introduced intothe field of sea clutter parameters estimation, and a reasonable model was built to make it have the high-precision parame?ters estimation ability of the sea clutter amplitude distribution model. The method uses the histogram statistics method topreprocess the data, rationalized the grouping interval of the input data labels, constructs the data set to training neural net?work, and use the test data to obtain the neural network estimation results. The simulation data and the X-band IPIX radarshow that compared with traditional mathematical statistical estimation method, the algorithm significantly improves the es?timation accuracy of the sea clutter statistical model parameter.