无人值守传感器目标分类方法

Target Classification Method for Unattended Sensors

摘要:

地面传感器阵列对目标的分类能力是影响其使用效能的关键因素之一。传感器阵列获取的观测数据受外部环境影响，当外部各种物理场特征随时间发生改变时，采用固定的分类标准进行判断会降低传感器阵列对目标分类识别结果的可信性。提出基于动态数据驱动的传感器阵列目标分类方法，对时变的外部环境特征加以提取作为目标识别计算的反馈控制量，构建环境改进模型作为决策用有限状态自动识别机，为其中每个状态训练相应的分类方法，实现对分类准则的自适应选择，提高算法性能。

Abstract:

Target classification capability using unattended ground sensors (UGS) has been a fatal factor for its operational performance. The sensor observations are significantly affected by external conditions. When the context is time-varying the usage of the same classifier may not be a good way to perform target classification. Dynamic Data Driven (DDD) method is proposed for dynamically extract and use the knowledge of context as feedback in order to adaptively choose the appropriate classifiers and thereby enhance the target classification performance. A context evolution model is constructed as Deterministic Finite State Automata (DFSA) and, for every context state in this DFSA, an event classifier is trained to classify the targets.