混沌信号在无线传感器网络中的盲分离

发布时间：2018-06-25 08:36

本文选题：混沌信号 + 盲分离　；参考：《物理学报》2014年05期

【摘要】：混沌信号在本质上属于非线性非高斯信号,它在无线传感器网络下的应用还涉及到信号量化问题,这使得混沌信号在此应用环境下的信号盲分离更为棘手.针对此问题,本文在容积卡尔曼粒子滤波的框架下提出一种解决方法.文中首先推导出观测信号的概率密度函数,在量化比特有限的情况下,采用最优量化器,获得最优的量化结果.在此基础上,使用容积卡尔曼滤波器产生粒子滤波中的重要性概率密度函数,融入最新的观测值,提高粒子对系统状态后验概率的逼近,提高信号盲分离的精度.仿真结果表明算法能够有效地分离混合混沌信号,参数估计的精度及其运算量均优于已有的无先导卡尔曼粒子滤波算法,其运行时间为无先导卡尔曼粒子滤波算法的88.77%.
[Abstract]:Chaotic signal is essentially a nonlinear non-Gao Si signal, and its application in wireless sensor networks also involves the problem of signal quantization, which makes the blind separation of chaotic signals in this application environment more difficult. In order to solve this problem, this paper presents a solution based on volumetric Kalman particle filter. In this paper, the probability density function of the observed signal is first derived. In the case of limited quantization bits, the optimal quantizer is used to obtain the optimal quantization results. On this basis, the volume Kalman filter is used to generate the important probability density function in particle filter, which integrates the latest observation values, improves the approximation of particle posteriori probability to the system state, and improves the accuracy of blind signal separation. The simulation results show that the proposed algorithm can effectively separate the mixed chaotic signals, and the precision and computational complexity of the parameter estimation are better than those of the existing unguided Kalman particle filter algorithm. The running time of the algorithm is 88.77g of the unleaded Kalman particle filter algorithm.
【作者单位】：华南理工大学电子与信息学院;
【基金】：国家自然科学基金(批准号:60872123,61101014) 国家-广东省自然科学基金联合基金(批准号:U0835001) 广东省高层次人才项目基金(批准号:N9101070) 中央高校基本业务费(批准号:2012ZM0025)资助的课题~~
【分类号】：TP212.91

【参考文献】