基于神经网络的多传感加权融合算法的研究
On the multi-sensor weighted fusion algorithm based on the neural network
云南民族大学学报:自然科学版,2016,25(1):75-80

简必建 JBJ

摘要


数据融合技术是无线传感器网络的一个关键的技术,能减少传感器的传输量,从而明显提高网络的感知性能,延长网络生命周期,减少时间延迟.多传感器的数据融合可以获得比单一传感器更多,更准确的信息.针对从含有噪声的测量数据中估计出监测变量,对自适应加权融合算法进行改进,基于信任度方法对测量数据进行数据预处理,然后基于神经网络误差修正的方法实现各传感器权重的自适应匹配,从而得到较为准确的估计值.通过对比仿真实验,算法的融合结果在精度、容错性方面均优于均值估计算法和自适应加权融合算法;能够更好地适应当今大数据环境下对数据精确度的要求. Data fusion technology is a key technology of wireless sensor networks, which can reduce the amount of transmission of the sensor data and the time delay, significantly improve the network's perceived performance, and extend its life cycle. Data fusion with multiple sensors can obtain more information than a single sensor. This research uses the variance estimation obtained from the noise-contained measurement data and then improves the adaptive weighted fusion algorithm. First of all, it preprocesses the measurement data with the CL method; then the adaptive matching of the weight of each sensor is realized based on the error correction of the neural network; finally it gets the effective estimate value. Simulation results show that the fusion results of the proposed algorithm are better than those of the mean value estimation algorithm and adaptive weighted fusion algorithm in precision and fault tolerance, which can better adapt to the requirements for data accuracy in today's big data environment.

参考



全文: PDF      下载: 652      浏览: 215


counter for myspace
云南民族大学学报(自然科学版) 1991—2016 Copyright
地址:云南省昆明市一二.一大街134号 邮编:650031 全国邮发代号:64-47
电话:0871-65132114 传真:0871-65137493 Email:ynmzxyxb@163.com