面向物联网的稀疏采样与近似重构技术研究
|
摘要
通过建立具有时空相关性的物联网无线传感器网络节点数据模型,对由数据样本构成的矩阵进行稀疏采样及矩阵填充的近似重构技术来实现数据的恢复,从而得到矩阵重构可行性(即采样比率和重构精度)与数据相关性的关系.仿真显示节点数据之间的时空相关性与矩阵重构可行性之间存在着密切的关系,同一采样比率条件下,时空相关性越大,重构精度越高;同样重构精度下,时空相关性越大,需要的采样比率越低.
In the report, the data model of wireless sensor network nodes with spatial-temporal correlation was established, the approximate reconstruction technique of sparse sampling and matrix filling was used to restore the data, and obtain the relationship between the feasibility of matrix reconstruction(i.e. sampling ratio and reconstruction accuracy) and the data correlation. The simulation results showed that there is a close relationship between the spatial-temporal correlation of node data and the feasibility of matrix reconstruction. Under the same sampling ratio, the greater the spatial-temporal correlation, the higher the reconstruction accuracy; under the same reconstruction accuracy, the greater the spatial-temporal correlation, the lower the sampling ratio required.
In the report, the data model of wireless sensor network nodes with spatial-temporal correlation was established, the approximate reconstruction technique of sparse sampling and matrix filling was used to restore the data, and obtain the relationship between the feasibility of matrix reconstruction(i.e. sampling ratio and reconstruction accuracy) and the data correlation. The simulation results showed that there is a close relationship between the spatial-temporal correlation of node data and the feasibility of matrix reconstruction. Under the same sampling ratio, the greater the spatial-temporal correlation, the higher the reconstruction accuracy; under the same reconstruction accuracy, the greater the spatial-temporal correlation, the lower the sampling ratio required.
引文
[1] 马磊.无线传感器网络在煤矿井下监测监控系统中的应用[J].无线互联科技,2015(17):22-23.
[2] 乔建华,张雪英.基于压缩感知的无线传感器网络数据收集研究综述[J].计算机应用,2017,37(11):3 261-3 269.
[3] 宋寿鹏,邵勇华,堵莹.采样方法研究综述[J].数据采集与处理,2016,31(3):452-463.
[4] 曹烁,刘志杰.基于非凸矩阵填充模型的图像修复方法研究[J].贵州师范大学学报(自然科学版),2018,36(3):89-94.
[5] 沈凤臣,王一舒,张龙辉.基于均匀空间采样模型的矩阵填充DOA估计方法[J].舰船电子对抗,2018,41(1):71-74.
[6] Zhang S Q,Dong G G,Kuang G Y.Matrix Completion for downward-looking 3-D SAR imaging with a random sparse[J].IEEE Transactions on Geoscience and Remote Sensing,2018,56(4):1 994-2 006.
[7] 杨建翠,李英.基于移动最小二乘法无线传感器网络数据压缩算法研究[J].科技通报,2018,34(1):176-179.
[8] 李腾非,魏璇,张怀相.基于压缩感知的无线传感器网络定位算法研究[J].杭州电子科技大学学报(自然科学版),2018,38(4):35-40.
[9] Wang K,Liu Y L,Wan Q,et al.Compressed sensing of wireless sensor networks data with missed measurements[J].Chinese Journal of Electronics,2015,24(2):388-392.
[10] Shen Y R,Hu W,Rana R,et al.Nonuniform compressive sensing for heterogeneous wireless sensor networks[J].IEEE Sensors Journal,2013,13(6):2 120-2 128.
[11] 吴桂峰,王轩.基于二次规划的无线传感器网络数据恢复算法[J].计算机应用,2013,33(4):935-938.
[12] 陈业斌,王仁伟,李颖.无线传感器网络中基于采样的时空数据恢复[J].重庆理工大学学报(自然科学),2017,31(6):127-133.
[13] Sun N,Wu J.Optimum sampling in a spatial-temporally correlated wireless sensor network:proceedings of the 2011 IEEE Global Telecommunications Conference,Kathmandu,December 5-9,2011[C].[S.l.]:IEEE,2012.
[14] 邹志强,王悦,沙超,等.无线传感器水下监测网络稀疏采样和近似重构[J].仪器仪表学报,2012,33(12):2 728-2 734.
[15] Eftekhari A,Yang D,Wakin M B.Weighted matrix completion and recovery with prior subspace information[J].IEEE Transactions on Information Theory,2018,64(6):4 044-4 071.
[16] 韦仙.基于矩阵填充技术重构低秩密度矩阵 [J].武汉工程大学学报,2015,37(2):72-76.
[17] 程晓军,江艺榕,潭忠.矩阵填充方法及其在高校综合排名中的应用[J].数学建模及其应用,2018,7(2):11-35.
[18] 王燕荣,尹佳杰,闫喜红.一种求解低秩矩阵填充问题的新方法[J].太原师范学院学报(自然科学版),2018,17(2):25-44.
[19] 王川龙,李晓丽.矩阵填充的子空间逼近法[J].中国科学:数学,2018,48(5):661-670.
[20] 张策,李鸥,童昕,等.基于压缩感知与矩阵补全技术的WSN数据收集算法[J].通信学报,2018,39(2):164-173.
[2] 乔建华,张雪英.基于压缩感知的无线传感器网络数据收集研究综述[J].计算机应用,2017,37(11):3 261-3 269.
[3] 宋寿鹏,邵勇华,堵莹.采样方法研究综述[J].数据采集与处理,2016,31(3):452-463.
[4] 曹烁,刘志杰.基于非凸矩阵填充模型的图像修复方法研究[J].贵州师范大学学报(自然科学版),2018,36(3):89-94.
[5] 沈凤臣,王一舒,张龙辉.基于均匀空间采样模型的矩阵填充DOA估计方法[J].舰船电子对抗,2018,41(1):71-74.
[6] Zhang S Q,Dong G G,Kuang G Y.Matrix Completion for downward-looking 3-D SAR imaging with a random sparse[J].IEEE Transactions on Geoscience and Remote Sensing,2018,56(4):1 994-2 006.
[7] 杨建翠,李英.基于移动最小二乘法无线传感器网络数据压缩算法研究[J].科技通报,2018,34(1):176-179.
[8] 李腾非,魏璇,张怀相.基于压缩感知的无线传感器网络定位算法研究[J].杭州电子科技大学学报(自然科学版),2018,38(4):35-40.
[9] Wang K,Liu Y L,Wan Q,et al.Compressed sensing of wireless sensor networks data with missed measurements[J].Chinese Journal of Electronics,2015,24(2):388-392.
[10] Shen Y R,Hu W,Rana R,et al.Nonuniform compressive sensing for heterogeneous wireless sensor networks[J].IEEE Sensors Journal,2013,13(6):2 120-2 128.
[11] 吴桂峰,王轩.基于二次规划的无线传感器网络数据恢复算法[J].计算机应用,2013,33(4):935-938.
[12] 陈业斌,王仁伟,李颖.无线传感器网络中基于采样的时空数据恢复[J].重庆理工大学学报(自然科学),2017,31(6):127-133.
[13] Sun N,Wu J.Optimum sampling in a spatial-temporally correlated wireless sensor network:proceedings of the 2011 IEEE Global Telecommunications Conference,Kathmandu,December 5-9,2011[C].[S.l.]:IEEE,2012.
[14] 邹志强,王悦,沙超,等.无线传感器水下监测网络稀疏采样和近似重构[J].仪器仪表学报,2012,33(12):2 728-2 734.
[15] Eftekhari A,Yang D,Wakin M B.Weighted matrix completion and recovery with prior subspace information[J].IEEE Transactions on Information Theory,2018,64(6):4 044-4 071.
[16] 韦仙.基于矩阵填充技术重构低秩密度矩阵 [J].武汉工程大学学报,2015,37(2):72-76.
[17] 程晓军,江艺榕,潭忠.矩阵填充方法及其在高校综合排名中的应用[J].数学建模及其应用,2018,7(2):11-35.
[18] 王燕荣,尹佳杰,闫喜红.一种求解低秩矩阵填充问题的新方法[J].太原师范学院学报(自然科学版),2018,17(2):25-44.
[19] 王川龙,李晓丽.矩阵填充的子空间逼近法[J].中国科学:数学,2018,48(5):661-670.
[20] 张策,李鸥,童昕,等.基于压缩感知与矩阵补全技术的WSN数据收集算法[J].通信学报,2018,39(2):164-173.