基于信息增量提出了一种多传感器对多目标检测与分类的优化算法。通过目标环境不确 定性定量描述的熵及其熵发生变化而产生的信息增量，给出了一种基于最大信息增量的传感 器对离散检测单元的搜索策略及算法实现。仿真结果表明，与顺序搜索方法相比，该算法大 大提高了信噪比，降低了错误率。%This research is a progress report on a project funded by NNSFC (National Natural Science Foundation of China). An optimizing algorithm based on information gain is put forward for multisensor detection and classification of multitargets. We deem that the uncertainty in target environment can be quantitatively described by entropy, and that we can use decrease in entropy to represent informationgain. In section 2 we discuss in much detail how to use entropy to obtain information gain and the optimizing algorithm of sensor management. Simulation results in Table 1 indicate preliminarily that, compared with sequence search algorithm , our optimizing algorithm can improve SNR (signal noise ratio) and reduce error probability.
展开▼
