基于联合卡尔曼滤波的异构网络数据信息融合方法  

Data Information Fusion of Heterogeneous Network Based on Combined Kalman Filtering

在线阅读下载全文

作  者:张凌志 ZHANG Lingzhi(Guangxi Gaomin Science and Technology Development Co:,LTD.WuzhouGuangxi 543000 China)

机构地区:[1]广西高民科技发展有限公司,广西梧州543000

出  处:《长江信息通信》2024年第5期13-15,共3页Changjiang Information & Communications

摘  要:信息融合是异构网络数据重要处理手段,现行方法融合效果并不理想,在实际应用中融合后数据值与真实值存在较大的差距,LSE值较大,针对现行方法存在的缺陷与不足提出基于联合卡尔曼滤波的异构网络数据信息融合方法。对异构网络数据中异常值识别,采用插值法对异常数据修复,对异构网络数据中野值识别与修正,并对异构网络数据时间配准,采用联合卡尔曼滤波算法估计异构网络数据最优值,消除数据采样误差与噪声,将不同数据源异构网络数据信息融合,以此实现了基于联合卡尔曼滤波的异构网络数据信息融合。实验证明,在设计方法应用下融合后数据值贴近于真实值,LSE值在0.1以下,具有较高的信息融合精度,在异构网络数据信息融合领域具有良好的应用前景。information fusion is an important means of heterogeneous network data,the current mcthod fusion effect is not ideal,in practical application after fusion data value and real value exist big gap,LSE value is larger,according to the current method of defects and deficiencies based on joint Kalman filter heterogencous nctwork data information fusion method.Outlier in heterogencous network data identification,the interpolation method for abnormal data repair,field in heterogeneous network data identification and correction,and the time registration of heterogeneous network data,using joint kalman filtering algorithm estimate heterogeneous net-work data optimal value,eliminate data sampling crror and noise,the different data source heter-ogencous network data information fusion,to realize the heterogeneous network data fusion based on combined data fusion kalman filter.It is proved that under the application of the design method,the fusion data value is close to the true value,and the LSE value is below o.1,which has a high information fusion accuracy,and has a good application prospect in the field of heter-ogeneous network data information fusion.

关 键 词:联合卡尔曼滤波 异构网络数据 信息融合 插值法 时间配准 LSE 

分 类 号:TN911[电子电信—通信与信息系统]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象