基于认知无线电系统合作检测的数据融合研究

Data fusion for cooperative detection in cognitive radio system

从多传感器数据融合角度研究认知无线电系统合作感知问题。对于特定的在线用户规模,存在一个最佳的参与部分融合用户数,可以固定选择每次检测的融和用户数使系统平均检测概率最大。在此基础上提出一种基于测量的融合方法,根据每次检测的接收信号状况动态调整参与融合的用户数。仿真表明,这种方法可以使系统检测性能在60%以上的检测中相对于固定融合用户数的方法有所提高。

The cooperative detection in cognitive radio system was investigated based on multi-sensor data fusion. For a certain scale of the online users, an optimal user number participating in data fusion existed. Based on this fact, a measure-based fusion method was proposed. This method dynamically adjusted the user number in fusion according to the received signal in each detection activity. Simulation shows the performance of over 60% detection is improved by this fusion method than the method with fixed fusion user number.