为了防止数据在开放的无线传感器网络中传输过程中遭到窃听而泄密,本文设计了一种带安全机制的传感器网络分布式决策融合方法 FRSM,该方法在传感节点发送数据到融合中心时利用随机函数对数据进行翻转置乱,以阻止窃听方获得正确的融合结果,而合法的融合中心能够基于已知的随机函数概率密度对接收到的数据进行对数似然比决策.理论分析和对比实验表明,FRSM能够使得合法的融合中心取得较好的融合结果,而窃听方在高信噪比和大量传感器环境下仍然难以获得正确的融合结果,从而保障了传感数据的机密性.
In the paper, the authors designed a distributed fusion rules with security scheme FRSM to prevent the eavesdropping during the data transmission in an open WSN. To make the enemy fusion cen- ter totally ignorant the true fusion result, the sensor data is randomly flipped by an random function with known probability density function, while the ally fusion center can properly detect an unknown target by performs the LLR fusion rules based on the known probability density function. The theory a- nalysis and comparison result demonstrated that good fusion result can be got by the AFC, and the EFC is prevent to detect target even under high signal-to-noise ratio or with lots of sensors, which guarantees the confidentiality of sensor data.