循环矩阵下压缩感知在瓦斯检测数据重构中的应用研究  

Application Research of Compressed Sensing in Gas Monitoring Data Reconstruction under Circular Matrix

作　　者：于瓅[1] 李世东 YU Li;LI Shidong(College of Computer Science and Engineering,Anhui University of Science and Technology,Huainan Anhui 232001,China;Department of Mathematics,San Francisco State University,San Francisco CA 94132,America)

机构地区：[1]安徽理工大学计算机科学与工程学院,安徽淮南232001 [2]旧金山州立大学数学系,旧金山加州94132

出　　处：《佳木斯大学学报（自然科学版）》2022年第4期148-152,共5页Journal of Jiamusi University：Natural Science Edition

基　　金：2021年安徽省重点研究与开发计划项目(202104d07020010)。

摘　　要：针对瓦斯监控预警系统中的传感器节点能量有限及传输带宽有限,为减少信息传输量和布置节点的数据量,采用基于零空间调整的重构方法对瓦斯数据进行重构。瓦斯传感器对应的观测矩阵是瓦斯传感器的脉冲响应函数循环而构成的循环矩阵,应用高斯函数序列构造的循环矩阵来模拟。选用不同煤矿的多组瓦斯监测数据作为实验数据,对瓦斯数据进行快速傅里叶变换,观测矩阵分别使用循环矩阵和随机矩阵。重构实验结果表明,模拟脉冲响应函数的循环矩阵对瓦斯检测信号重构效果较好,当采样率一定范围减少时,循环矩阵下重构的平均误差约5%。因此,实际工程中可由降低50%采样率的数据来重构原始数据,显著减少数据采样点数,节省工程检测成本。一定范围内可用低分辨率的瓦斯传感器代替高分辨率的瓦斯传感器,以降低构建瓦斯监控物联网的成本。In view of the limited energy and transmission bandwidth of sensor nodes in gas monitoring and early warning system,in order to reduce the amount of information transmission and the amount of data arranged at nodes,the reconstruction method based on zero space adjustment is used to reconstruct gas data.The observation matrix corresponding to the gas sensor is a cyclic matrix formed by the cyclic impulse response function of the gas sensor,which is simulated by the cyclic matrix constructed by Gaussian function sequence.Several groups of gas monitoring data from different coal mines are selected as experimental data,and the gas data are transformed by fast Fourier transform.The observation matrix uses cyclic matrix and random matrix respectively.The reconstruction experimental results show that the cyclic matrix of analog impulse response function has a good effect on the reconstruction of gas detection signal.When the sampling rate is reduced in a certain range,the average error of reconstruction under the cyclic matrix is about 5%.Therefore,in practical engineering,the original data can be reconstructed by reducing the sampling rate by 50%,which can significantly reduce the number of data sampling points and save the engineering detection cost.Low resolution gas sensor can be used to replace high-resolution gas sensor in a certain range,so as to reduce the cost of building gas monitoring Internet of things.

关 键 词：瓦斯监控物联网 稀疏分析 瓦斯数据重构 高斯函数序列 循环矩阵 随机矩阵 

分 类 号：X9[环境科学与工程—安全科学]