基于类决策树分类的特征层融合识别算法

针对雷达组网量测数据不确定性大、信息不完备等特点, 基于决策树分类算法的思想, 创建类决策树的概念, 提出一种基于类决策树分类的特征层融合识别算法. 所给出的算法无需训练样本, 采用边构造边分类的方式, 选取信 息增益最大的属性作为分类属性对量测数据进行分类, 实现了对目标的识别. 该算法能够处理含有空缺值的量测数据, 充分利用量测数据的特征信息. 仿真实验结果表明, 类决策树分类算法是一种简单有效的特征层融合识别算法.

Feature-level fusion recognition algorithm based on analogy decision tree classification

Considering the big uncertainty and incomplete information of radar network measurement data, the concept of analogy decision tree is created and a feature-level fusion recognition algorithm based on decision tree analogy is proposed. The proposed algorithm uses a way of while constructing while classifying without training samples. The greatest information gain property feature is selected as the classification feature to classify the measurement data, which achieves the goal of recognizing the target. The algorithm can deal with the measurement data containing vacancies and makes full use of measurement data. Simulation results show that the analogy decision tree classification algorithm is a simple and effective feature-level fusion recognition algorithm.