在分析径向基函数网络（ＲａｄｉａｌＢａｓｉｓＦｕｎｃｔｉｏｎＮｅｔｗｏｒｋ，ＲＢＦＮ）的基础上，提出了一种更适合于目标识别的基于模糊聚类的径向基函数网络（ＦｕｚｚｙＣｌｕｓｔｅｒｉｎｇＢａｓｅｄＲａｄｉａｌＢａｓｉｓＦｕｎｃｔｉｏｎＮｅｔｗｏｒｋ，ＦＣＢＲＢＦＮ）。这种网络利用模糊聚类方法，根据训练样本的空间分布确定网络的结构，利用聚类结果中的隶属度函数值控制高斯核函数形状参数。理论分析还表明，此径向基函数网络具有比一般径向基函数网络更强的泛化能力。利用一种战场侦察雷达获取的回波数据进行实验，结果表明，基于模糊聚类的该径向基函数网络的分类结果优于一般径向基函数网络。

The RBFN (Radial Basis Function Network) is analyzed and the FCBRBFN(Fuzzy Clustering Based Radial Basis Function Network) which is more suitable for radar target recognition is proposed in this paper. This FCBRBFN utilizes fuzzy clustering method to determine the structure of the net. It estimates the shape parameter of Guassian kernel function with the values of membership function which gets from the clustering results. It is shown from the theoretical analysis that the FCBRBFN has better generalization ability. The Doppler echoes of the targets gotten from a current surveillance radar are used in the experiment. The experimental results show that the classification rate of the FCBRBFN is higher than that of the RBFN. The network proposed in this paper is promising in the application of radar target recognition.

TN959.17

国家博士点基金