基于扩散式信息融合的空间分布式变阶数自适应网络滤波方法研究

随着传感器网络技术的发展，空间分布式自适应网络滤波方法以其较高的估计精度得到了人们的关注。然而现有的空间分布式自适应网络滤波方法只能在系统阶数已知条件下应用，且存在收敛速度慢、鲁棒性较差等问题，影响了其广泛应用。为解决上述问题，本课题拟展开基于扩散式信息融合的空间分布式变阶数自适应网络滤波方法研究，通过扩散式信息融合方法自适应迭代寻求最优滤波器阶数，解决空间分布式自适应网络滤波方法在系统阶数未知情况下的应用问题；采用不等式约束思想，通过优化系统参数解决空间分布式自适应网络滤波方法快速收敛问题；通过节点性能检测和故障隔离，解决空间分布式自适应网络滤波方法对噪声干扰的鲁棒性问题。本项目的研究成果将丰富自适应网络滤波方法基础理论研究，并促进自适应网络滤波方法的广泛应用。

With the development of sensor network technologies, spatio distributed adaptive network filtering algorithm has attracted focus because of its high accuracy. However, current spatio distributed adaptive network filtering algorithm can only be used when the tap-length of unknown system is known. It also has disadvantages of slow convergence rate and non-robust to interference, which limits its applications. To solve the above problems, research on variable tap-length distributed adaptive network filtering algorithm based on spatio diffusion will be performed in this project. In this method, adaptive diffusion method will be used to search for the optimal tap-length, which can solve the problem of unknown system tap-length; inequality constraint method will be used to optimize parameters to solve the problem of fast convergence; node performance detection and fault isolation method will be used to solve the problem of robust to interference. Research results of this project will be meaningful to adaptive network filtering theory and facilitate the application of adaptive network filtering algorithm.