基于一种改进的压缩感知信号重建算法在WSN中的研究

ResearchSbasedSonSanSImprovedSCompressionSSensingSSignalSReconstructionSAlgorithmSinSWSN

摘要:

无线传感中的数据节点能量消耗一直都是研究的重点,本文在WSN的数据压缩上提出一种提近似L0范数的压缩感知信号重构算法,本算法不需要将信号稀疏作为前提条件,而是采用反双曲线正弦函数来对估计有关SL0范数的最小的值,采用牛顿-割线法来计算步长因子,求解使得算法收敛速度更快,并在一定程度上避免“波浪现象”；提高了信号的重建质量,降低了重构网络数据所需要的测量数。仿真实验表明本文的算法与现有的WSN压缩算法相比,能够有效的减少数据传输过程中的数据量,进一步降低网络节点消耗。

Abstract:

TheSenergySconsumptionSofSdataSnodesSinStheSwirelessSsensorShasSalwaysSbeenSaSfocusSofSresearch.SThisSpaperSproposesSaScompressedSsensingSsignalSreconstructionSalgorithmSapproximatingSL0SnormSbasedSonSWSNSdataScompression.SSignalSsparsenessSisSnotSneededStoSbeSasSaSprerequisiteSinSthisSalgorithm,Sinstead,SinverseShyperbolicSsineSfunctionSisSadoptedStoSestimateStheSminimumSvalueSofSSL0Snorm,SandSNewtonS-SsecantSmethodSisSusedStoScalculateStheSstepSfactor,SthusSmakingStheSalgorithmSconvergeSfasterSandSavoidingS“waveSphenomenon”StoSsomeSextent;SimprovingStheSsignalSreconstructionSquality,SandSreducingStheSmeasurementSnumberSrequiredSforSnetworkSdataSreconstruction.SSimulationSexperimentsSshowSthatScomparedSwithStheSexistingSWSNScompressionSalgorithm,SalgorithmSinSthisSpaperScanSeffectivelySreduceStheSdataSamountSduringSdataStransmissionSprocess,SandSfurtherSreduceSnetworkSnodeSconsumption.