论文部分内容阅读
为减少基于似然比检测的无线传感器网络最优分布式检测方法对信道状态、噪声分布和传感器检测性能等信息的依赖,满足实际系统的需要,提出使用支持向量机实现无线传感器网络分布式信号检测的方法.该方法无需预知信道状态和噪声分布信息,简化了无线传感器的节点功能.该方法将传感器节点测量结果以模拟非编码方式发送至数据中心,数据中心利用训练好的支持向量机对接收到的信号进行判决.仿真结果表明,该方法可获得较高的检测精度.
In order to reduce the dependence on the channel state, noise distribution and sensor detection performance of the optimal distributed detection method of wireless sensor network based on likelihood ratio detection to meet the needs of the actual system, this paper proposes to use support vector machine to realize the distributed signal of wireless sensor network This method simplifies the sensor function of the wireless sensor without knowing the channel state and noise distribution information.The method sends the measurement result of the sensor node to the data center by analog non-coding method, and the data center utilizes the trained support vector machine The received signal is judged.The simulation results show that the method can obtain higher detection accuracy.